Machine Learning
See recent articles
Showing new listings for Friday, 11 April 2025
- [1] arXiv:2504.07151 [pdf, other]
-
Title: Deep Sturm--Liouville: From Sample-Based to 1D Regularization with Learnable Orthogonal Basis FunctionsSubjects: Machine Learning (cs.LG)
Although Artificial Neural Networks (ANNs) have achieved remarkable success across various tasks, they still suffer from limited generalization. We hypothesize that this limitation arises from the traditional sample-based (0--dimensionnal) regularization used in ANNs. To overcome this, we introduce \textit{Deep Sturm--Liouville} (DSL), a novel function approximator that enables continuous 1D regularization along field lines in the input space by integrating the Sturm--Liouville Theorem (SLT) into the deep learning framework. DSL defines field lines traversing the input space, along which a Sturm--Liouville problem is solved to generate orthogonal basis functions, enforcing implicit regularization thanks to the desirable properties of SLT. These basis functions are linearly combined to construct the DSL approximator. Both the vector field and basis functions are parameterized by neural networks and learned jointly. We demonstrate that the DSL formulation naturally arises when solving a Rank-1 Parabolic Eigenvalue Problem. DSL is trained efficiently using stochastic gradient descent via implicit differentiation. DSL achieves competitive performance and demonstrate improved sample efficiency on diverse multivariate datasets including high-dimensional image datasets such as MNIST and CIFAR-10.
- [2] arXiv:2504.07155 [pdf, html, other]
-
Title: Compound Fault Diagnosis for Train Transmission Systems Using Deep Learning with Fourier-enhanced RepresentationComments: Accepted for the 2025 IEEE Conference on Prognostics and Health Management (ICPHM 2025)Subjects: Machine Learning (cs.LG)
Fault diagnosis prevents train disruptions by ensuring the stability and reliability of their transmission systems. Data-driven fault diagnosis models have several advantages over traditional methods in terms of dealing with non-linearity, adaptability, scalability, and automation. However, existing data-driven models are trained on separate transmission components and only consider single faults due to the limitations of existing datasets. These models will perform worse in scenarios where components operate with each other at the same time, affecting each component's vibration signals. To address some of these challenges, we propose a frequency domain representation and a 1-dimensional convolutional neural network for compound fault diagnosis and applied it on the PHM Beijing 2024 dataset, which includes 21 sensor channels, 17 single faults, and 42 compound faults from 4 interacting components, that is, motor, gearbox, left axle box, and right axle box. Our proposed model achieved 97.67% and 93.93% accuracies on the test set with 17 single faults and on the test set with 42 compound faults, respectively.
- [3] arXiv:2504.07158 [pdf, html, other]
-
Title: Holistic Capability Preservation: Towards Compact Yet Comprehensive Reasoning ModelsLing Team: Caizhi Tang, Chilin Fu, Chunwei Wu, Jia Guo, Jianwen Wang, Jingyu Hu, Liang Jiang, Meng Li, Peng Jiao, Pingping Liu, Shaomian Zheng, Shiwei Liang, Shuaicheng Li, Yalin Zhang, Yingting Wu, Yongkang Liu, Zhenyu HuangComments: 10 pagesSubjects: Machine Learning (cs.LG); Computation and Language (cs.CL)
This technical report presents Ring-Lite-Distill, a lightweight reasoning model derived from our open-source Mixture-of-Experts (MoE) Large Language Models (LLMs) Ling-Lite. This study demonstrates that through meticulous high-quality data curation and ingenious training paradigms, the compact MoE model Ling-Lite can be further trained to achieve exceptional reasoning capabilities, while maintaining its parameter-efficient architecture with only 2.75 billion activated parameters, establishing an efficient lightweight reasoning architecture. In particular, in constructing this model, we have not merely focused on enhancing advanced reasoning capabilities, exemplified by high-difficulty mathematical problem solving, but rather aimed to develop a reasoning model with more comprehensive competency coverage. Our approach ensures coverage across reasoning tasks of varying difficulty levels while preserving generic capabilities, such as instruction following, tool use, and knowledge retention. We show that, Ring-Lite-Distill's reasoning ability reaches a level comparable to DeepSeek-R1-Distill-Qwen-7B, while its general capabilities significantly surpass those of DeepSeek-R1-Distill-Qwen-7B. The models are accessible at this https URL
- [4] arXiv:2504.07170 [pdf, html, other]
-
Title: Trustworthy AI Must Account for IntersectionalityComments: Presented at the ICLR 2025 Workshop on Bidirectional Human-AI AlignmentSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Trustworthy AI encompasses many aspirational aspects for aligning AI systems with human values, including fairness, privacy, robustness, explainability, and uncertainty quantification. However, efforts to enhance one aspect often introduce unintended trade-offs that negatively impact others, making it challenging to improve all aspects simultaneously. In this position paper, we review notable approaches to these five aspects and systematically consider every pair, detailing the negative interactions that can arise. For example, applying differential privacy to model training can amplify biases in the data, undermining fairness. Drawing on these findings, we take the position that addressing trustworthiness along each axis in isolation is insufficient. Instead, research on Trustworthy AI must account for intersectionality between aspects and adopt a holistic view across all relevant axes at once. To illustrate our perspective, we provide guidance on how researchers can work towards integrated trustworthiness, a case study on how intersectionality applies to the financial industry, and alternative views to our position.
- [5] arXiv:2504.07240 [pdf, html, other]
-
Title: Prototype-Based Continual Learning with Label-free Replay Buffer and Cluster Preservation LossSubjects: Machine Learning (cs.LG)
Continual learning techniques employ simple replay sample selection processes and use them during subsequent tasks. Typically, they rely on labeled data. In this paper, we depart from this by automatically selecting prototypes stored without labels, preserving cluster structures in the latent space across tasks. By eliminating label dependence in the replay buffer and introducing cluster preservation loss, it is demonstrated that the proposed method can maintain essential information from previously encountered tasks while ensuring adaptation to new tasks. "Push-away" and "pull-toward" mechanisms over previously learned prototypes are also introduced for class-incremental and domain-incremental scenarios. These mechanisms ensure the retention of previously learned information as well as adaptation to new classes or domain shifts. The proposed method is evaluated on several benchmarks, including SplitCIFAR100, SplitImageNet32, SplitTinyImageNet, and SplitCaltech256 for class-incremental, as well as R-MNIST and CORe50 for domain-incremental setting using pre-extracted DINOv2 features. Experimental results indicate that the label-free replay-based technique outperforms state-of-the-art continual learning methods and, in some cases, even surpasses offline learning. An unsupervised variant of the proposed technique for the class-incremental setting, avoiding labels use even on incoming data, also demonstrated competitive performance, outperforming particular supervised baselines in some cases. These findings underscore the effectiveness of the proposed framework in retaining prior information and facilitating continual adaptation.
- [6] arXiv:2504.07247 [pdf, html, other]
-
Title: Resource-efficient Inference with Foundation Model ProgramsSubjects: Machine Learning (cs.LG)
The inference-time resource costs of large language and vision models present a growing challenge in production deployments. We propose the use of foundation model programs, i.e., programs that can invoke foundation models with varying resource costs and performance, as an approach to this problem. Specifically, we present a method that translates a task into a program, then learns a policy for resource allocation that, on each input, selects foundation model "backends" for each program module. The policy uses smaller, cheaper backends to handle simpler subtasks, while allowing more complex subtasks to leverage larger, more capable models. We evaluate the method on two new "streaming" visual question-answering tasks in which a system answers a question on a sequence of inputs, receiving ground-truth feedback after each answer. Compared to monolithic multi-modal models, our implementation achieves up to 98% resource savings with minimal accuracy loss, demonstrating its potential for scalable and resource-efficient multi-modal inference.
- [7] arXiv:2504.07261 [pdf, html, other]
-
Title: Adapting to Online Distribution Shifts in Deep Learning: A Black-Box ApproachComments: To appear at AISTATS 2025Subjects: Machine Learning (cs.LG)
We study the well-motivated problem of online distribution shift in which the data arrive in batches and the distribution of each batch can change arbitrarily over time. Since the shifts can be large or small, abrupt or gradual, the length of the relevant historical data to learn from may vary over time, which poses a major challenge in designing algorithms that can automatically adapt to the best ``attention span'' while remaining computationally efficient. We propose a meta-algorithm that takes any network architecture and any Online Learner (OL) algorithm as input and produces a new algorithm which provably enhances the performance of the given OL under non-stationarity. Our algorithm is efficient (it requires maintaining only $O(\log(T))$ OL instances) and adaptive (it automatically chooses OL instances with the ideal ``attention'' length at every timestamp). Experiments on various real-world datasets across text and image modalities show that our method consistently improves the accuracy of user specified OL algorithms for classification tasks. Key novel algorithmic ingredients include a \emph{multi-resolution instance} design inspired by wavelet theory and a cross-validation-through-time technique. Both could be of independent interest.
- [8] arXiv:2504.07278 [pdf, other]
-
Title: A Multi-Phase Analysis of Blood Culture Stewardship: Machine Learning Prediction, Expert Recommendation Assessment, and LLM AutomationFatemeh Amrollahi, Nicholas Marshall, Fateme Nateghi Haredasht, Kameron C Black, Aydin Zahedivash, Manoj V Maddali, Stephen P. Ma, Amy Chang, MD Phar Stanley C Deresinski, Mary Kane Goldstein, Steven M. Asch, Niaz Banaei, Jonathan H ChenComments: 10 pages, 2 figures, 2 tables, conferenceSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Blood cultures are often over ordered without clear justification, straining healthcare resources and contributing to inappropriate antibiotic use pressures worsened by the global shortage. In study of 135483 emergency department (ED) blood culture orders, we developed machine learning (ML) models to predict the risk of bacteremia using structured electronic health record (EHR) data and provider notes via a large language model (LLM). The structured models AUC improved from 0.76 to 0.79 with note embeddings and reached 0.81 with added diagnosis codes. Compared to an expert recommendation framework applied by human reviewers and an LLM-based pipeline, our ML approach offered higher specificity without compromising sensitivity. The recommendation framework achieved sensitivity 86%, specificity 57%, while the LLM maintained high sensitivity (96%) but over classified negatives, reducing specificity (16%). These findings demonstrate that ML models integrating structured and unstructured data can outperform consensus recommendations, enhancing diagnostic stewardship beyond existing standards of care.
- [9] arXiv:2504.07297 [pdf, other]
-
Title: Data Fusion of Deep Learned Molecular Embeddings for Property PredictionSubjects: Machine Learning (cs.LG); Materials Science (cond-mat.mtrl-sci)
Data-driven approaches such as deep learning can result in predictive models for material properties with exceptional accuracy and efficiency. However, in many problems data is sparse, severely limiting their accuracy and applicability. To improve predictions, techniques such as transfer learning and multi-task learning have been used. The performance of multi-task learning models depends on the strength of the underlying correlations between tasks and the completeness of the dataset. We find that standard multi-task models tend to underperform when trained on sparse datasets with weakly correlated properties. To address this gap, we use data fusion techniques to combine the learned molecular embeddings of various single-task models and trained a multi-task model on this combined embedding. We apply this technique to a widely used benchmark dataset of quantum chemistry data for small molecules as well as a newly compiled sparse dataset of experimental data collected from literature and our own quantum chemistry and thermochemical calculations. The results show that the fused, multi-task models outperform standard multi-task models for sparse datasets and can provide enhanced prediction on data-limited properties compared to single-task models.
- [10] arXiv:2504.07307 [pdf, html, other]
-
Title: Follow-the-Perturbed-Leader Achieves Best-of-Both-Worlds for the m-Set Semi-Bandit ProblemsSubjects: Machine Learning (cs.LG); Machine Learning (stat.ML)
We consider a common case of the combinatorial semi-bandit problem, the $m$-set semi-bandit, where the learner exactly selects $m$ arms from the total $d$ arms. In the adversarial setting, the best regret bound, known to be $\mathcal{O}(\sqrt{nmd})$ for time horizon $n$, is achieved by the well-known Follow-the-Regularized-Leader (FTRL) policy, which, however, requires to explicitly compute the arm-selection probabilities by solving optimizing problems at each time step and sample according to it. This problem can be avoided by the Follow-the-Perturbed-Leader (FTPL) policy, which simply pulls the $m$ arms that rank among the $m$ smallest (estimated) loss with random perturbation. In this paper, we show that FTPL with a Fréchet perturbation also enjoys the optimal regret bound $\mathcal{O}(\sqrt{nmd})$ in the adversarial setting and achieves best-of-both-world regret bounds, i.e., achieves a logarithmic regret for the stochastic setting.
- [11] arXiv:2504.07322 [pdf, html, other]
-
Title: Bregman-Hausdorff divergence: strengthening the connections between computational geometry and machine learningComments: 23 pages, 11 figures, 3 tables, 3 algorithms, submitted to Machine Learning and Knowledge ExtractionSubjects: Machine Learning (cs.LG); Computational Geometry (cs.CG); Information Theory (cs.IT)
The purpose of this paper is twofold. On a technical side, we propose an extension of the Hausdorff distance from metric spaces to spaces equipped with asymmetric distance measures. Specifically, we focus on the family of Bregman divergences, which includes the popular Kullback--Leibler divergence (also known as relative entropy).
As a proof of concept, we use the resulting Bregman--Hausdorff divergence to compare two collections of probabilistic predictions produced by different machine learning models trained using the relative entropy loss. The algorithms we propose are surprisingly efficient even for large inputs with hundreds of dimensions.
In addition to the introduction of this technical concept, we provide a survey. It outlines the basics of Bregman geometry, as well as computational geometry algorithms. We focus on algorithms that are compatible with this geometry and are relevant for machine learning. - [12] arXiv:2504.07337 [pdf, html, other]
-
Title: FLASH: Flexible Learning of Adaptive Sampling from History in Temporal Graph Neural NetworksComments: 22 pages, 4 figures, 12 tablesSubjects: Machine Learning (cs.LG); Social and Information Networks (cs.SI)
Aggregating temporal signals from historic interactions is a key step in future link prediction on dynamic graphs. However, incorporating long histories is resource-intensive. Hence, temporal graph neural networks (TGNNs) often rely on historical neighbors sampling heuristics such as uniform sampling or recent neighbors selection. These heuristics are static and fail to adapt to the underlying graph structure. We introduce FLASH, a learnable and graph-adaptive neighborhood selection mechanism that generalizes existing heuristics. FLASH integrates seamlessly into TGNNs and is trained end-to-end using a self-supervised ranking loss. We provide theoretical evidence that commonly used heuristics hinders TGNNs performance, motivating our design. Extensive experiments across multiple benchmarks demonstrate consistent and significant performance improvements for TGNNs equipped with FLASH.
- [13] arXiv:2504.07342 [pdf, other]
-
Title: Leveraging deep learning for plant disease identification: a bibliometric analysis in SCOPUS from 2018 to 2024Journal-ref: Journal of Scientific Agriculture, 2025Subjects: Machine Learning (cs.LG)
This work aimed to present a bibliometric analysis of deep learning research for plant disease identification, with a special focus on generative modeling. A thorough analysis of SCOPUS-sourced bibliometric data from 253 documents was performed. Key performance metrics such as accuracy, precision, recall, and F1-score were analyzed for generative modeling. The findings highlighted significant contributions from some authors Too and Arnal Barbedo, whose works had notable citation counts, suggesting their influence on the academic community. Co-authorship networks revealed strong collaborative clusters, while keyword analysis identified emerging research gaps. This study highlights the role of collaboration and citation metrics in shaping research directions and enhancing the impact of scholarly work in applications of deep learning to plant disease identification. Future research should explore the methodologies of highly cited studies to inform best practices and policy-making.
- [14] arXiv:2504.07371 [pdf, html, other]
-
Title: Minimum width for universal approximation using squashable activation functionsSubjects: Machine Learning (cs.LG)
The exact minimum width that allows for universal approximation of unbounded-depth networks is known only for ReLU and its variants. In this work, we study the minimum width of networks using general activation functions. Specifically, we focus on squashable functions that can approximate the identity function and binary step function by alternatively composing with affine transformations. We show that for networks using a squashable activation function to universally approximate $L^p$ functions from $[0,1]^{d_x}$ to $\mathbb R^{d_y}$, the minimum width is $\max\{d_x,d_y,2\}$ unless $d_x=d_y=1$; the same bound holds for $d_x=d_y=1$ if the activation function is monotone. We then provide sufficient conditions for squashability and show that all non-affine analytic functions and a class of piecewise functions are squashable, i.e., our minimum width result holds for those general classes of activation functions.
- [15] arXiv:2504.07373 [pdf, html, other]
-
Title: ChronoFormer: Time-Aware Transformer Architectures for Structured Clinical Event ModelingSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
The temporal complexity of electronic health record (EHR) data presents significant challenges for predicting clinical outcomes using machine learning. This paper proposes ChronoFormer, an innovative transformer based architecture specifically designed to encode and leverage temporal dependencies in longitudinal patient data. ChronoFormer integrates temporal embeddings, hierarchical attention mechanisms, and domain specific masking techniques. Extensive experiments conducted on three benchmark tasks mortality prediction, readmission prediction, and long term comorbidity onset demonstrate substantial improvements over current state of the art methods. Furthermore, detailed analyses of attention patterns underscore ChronoFormer's capability to capture clinically meaningful long range temporal relationships.
- [16] arXiv:2504.07383 [pdf, html, other]
-
Title: PROPEL: Supervised and Reinforcement Learning for Large-Scale Supply Chain PlanningSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
This paper considers how to fuse Machine Learning (ML) and optimization to solve large-scale Supply Chain Planning (SCP) optimization problems. These problems can be formulated as MIP models which feature both integer (non-binary) and continuous variables, as well as flow balance and capacity constraints. This raises fundamental challenges for existing integrations of ML and optimization that have focused on binary MIPs and graph problems. To address these, the paper proposes PROPEL, a new framework that combines optimization with both supervised and Deep Reinforcement Learning (DRL) to reduce the size of search space significantly. PROPEL uses supervised learning, not to predict the values of all integer variables, but to identify the variables that are fixed to zero in the optimal solution, leveraging the structure of SCP applications. PROPEL includes a DRL component that selects which fixed-at-zero variables must be relaxed to improve solution quality when the supervised learning step does not produce a solution with the desired optimality tolerance. PROPEL has been applied to industrial supply chain planning optimizations with millions of variables. The computational results show dramatic improvements in solution times and quality, including a 60% reduction in primal integral and an 88% primal gap reduction, and improvement factors of up to 13.57 and 15.92, respectively.
- [17] arXiv:2504.07389 [pdf, html, other]
-
Title: Task-Circuit Quantization: Leveraging Knowledge Localization and Interpretability for CompressionComments: 24 pages. Code: this https URLSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
Post-training quantization (PTQ) reduces a model's memory footprint by mapping full precision weights into low bit weights without costly retraining, but can degrade its downstream performance especially in low 2- to 3-bit settings. We develop a new mixed-precision PTQ approach, Task-Circuit Quantization (TaCQ), that draws parallels to automated circuit discovery, directly conditioning the quantization process on specific weight circuits -- which we define as sets of weights associated with downstream task performance. These weights are kept as 16-bit weights, while others are quantized, maintaining performance while only adding a marginal memory cost. Specifically, TaCQ contrasts unquantized model weights with a uniformly-quantized model to estimate the expected change in weights due to quantization and uses gradient information to predict the resulting impact on task performance, allowing us to preserve task-specific weights. We compare TaCQ-based quantization to existing mixed-precision quantization methods when conditioning both on general-purpose and task-specific data. Across QA, math reasoning, and text-to-SQL tasks for both Llama-3 and Qwen2.5, we find that TaCQ outperforms baselines using the same calibration data and a lower weight budget, achieving major improvements in the 2 and 3-bit regime. With only 3.1 bits we are able to recover 96% of Llama-3-8B-Instruct's unquantized 16-bit MMLU performance, obtaining a 5.25% absolute improvement over SPQR. We also observe consistently large gains over existing methods in the 2-bit regime, with an average gain of 14.74% over the strongest baseline, SliM-LLM. Moreover, we observe a 7.20% gain without conditioning on specific tasks, showing TaCQ's ability to identify important weights is not limited to task-conditioned settings.
- [18] arXiv:2504.07393 [pdf, html, other]
-
Title: State Estimation Using Particle Filtering in Adaptive Machine Learning Methods: Integrating Q-Learning and NEAT Algorithms with Noisy Radar MeasurementsSubjects: Machine Learning (cs.LG); Optimization and Control (math.OC)
Reliable state estimation is essential for autonomous systems operating in complex, noisy environments. Classical filtering approaches, such as the Kalman filter, can struggle when facing nonlinear dynamics or non-Gaussian noise, and even more flexible particle filters often encounter sample degeneracy or high computational costs in large-scale domains. Meanwhile, adaptive machine learning techniques, including Q-learning and neuroevolutionary algorithms such as NEAT, rely heavily on accurate state feedback to guide learning; when sensor data are imperfect, these methods suffer from degraded convergence and suboptimal performance. In this paper, we propose an integrated framework that unifies particle filtering with Q-learning and NEAT to explicitly address the challenge of noisy measurements. By refining radar-based observations into reliable state estimates, our particle filter drives more stable policy updates (in Q-learning) or controller evolution (in NEAT), allowing both reinforcement learning and neuroevolution to converge faster, achieve higher returns or fitness, and exhibit greater resilience to sensor uncertainty. Experiments on grid-based navigation and a simulated car environment highlight consistent gains in training stability, final performance, and success rates over baselines lacking advanced filtering. Altogether, these findings underscore that accurate state estimation is not merely a preprocessing step, but a vital component capable of substantially enhancing adaptive machine learning in real-world applications plagued by sensor noise.
- [19] arXiv:2504.07394 [pdf, html, other]
-
Title: ClimateBench-M: A Multi-Modal Climate Data Benchmark with a Simple Generative MethodDongqi Fu, Yada Zhu, Zhining Liu, Lecheng Zheng, Xiao Lin, Zihao Li, Liri Fang, Katherine Tieu, Onkar Bhardwaj, Kommy Weldemariam, Hanghang Tong, Hendrik Hamann, Jingrui HeComments: Preprint, 29 pagesSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Climate science studies the structure and dynamics of Earth's climate system and seeks to understand how climate changes over time, where the data is usually stored in the format of time series, recording the climate features, geolocation, time attributes, etc. Recently, much research attention has been paid to the climate benchmarks. In addition to the most common task of weather forecasting, several pioneering benchmark works are proposed for extending the modality, such as domain-specific applications like tropical cyclone intensity prediction and flash flood damage estimation, or climate statement and confidence level in the format of natural language. To further motivate the artificial general intelligence development for climate science, in this paper, we first contribute a multi-modal climate benchmark, i.e., ClimateBench-M, which aligns (1) the time series climate data from ERA5, (2) extreme weather events data from NOAA, and (3) satellite image data from NASA HLS based on a unified spatial-temporal granularity. Second, under each data modality, we also propose a simple but strong generative method that could produce competitive performance in weather forecasting, thunderstorm alerts, and crop segmentation tasks in the proposed ClimateBench-M. The data and code of ClimateBench-M are publicly available at this https URL.
- [20] arXiv:2504.07397 [pdf, html, other]
-
Title: MicroNAS: An Automated Framework for Developing a Fall Detection SystemSeyed Mojtaba Mohasel, John Sheppard, Lindsey K. Molina, Richard R. Neptune, Shane R. Wurdeman, Corey A. PewSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
This work presents MicroNAS, an automated neural architecture search tool specifically designed to create models optimized for microcontrollers with small memory resources. The ESP32 microcontroller, with 320 KB of memory, is used as the target platform. The artificial intelligence contribution lies in a novel method for optimizing convolutional neural network and gated recurrent unit architectures by considering the memory size of the target microcontroller as a guide. A comparison is made between memory-driven model optimization and traditional two-stage methods, which use pruning, to show the effectiveness of the proposed framework. To demonstrate the engineering application of MicroNAS, a fall detection system (FDS) for lower-limb amputees is developed as a pilot study. A critical challenge in fall detection studies, class imbalance in the dataset, is addressed. The results show that MicroNAS models achieved higher F1-scores than alternative approaches, such as ensemble methods and H2O Automated Machine Learning, presenting a significant step forward in real-time FDS development. Biomechanists using body-worn sensors for activity detection can adopt the open-source code to design machine learning models tailored for microcontroller platforms with limited memory.
- [21] arXiv:2504.07402 [pdf, html, other]
-
Title: LauraTSE: Target Speaker Extraction using Auto-Regressive Decoder-Only Language ModelsComments: 5 pages, 1 figureSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
We propose LauraTSE, an Auto-Regressive Decoder-Only Language Model for Target Speaker Extraction (TSE) based on the LauraGPT backbone. It employs a small-scale auto-regressive decoder-only language model which takes the continuous representations for both the mixture and the reference speeches and produces the first few layers of the target speech's discrete codec representations. In addition, a one-step encoder-only language model reconstructs the sum of the predicted codec embeddings using both the mixture and the reference information. Our approach achieves superior or comparable performance to existing generative and discriminative TSE models. To the best of our knowledge, LauraTSE is the first single-task TSE model to leverage an auto-regressive decoder-only language model as the backbone.
- [22] arXiv:2504.07403 [pdf, html, other]
-
Title: Multi-Selection for Recommendation SystemsSubjects: Machine Learning (cs.LG)
We present the construction of a multi-selection model to answer differentially private queries in the context of recommendation systems. The server sends back multiple recommendations and a ``local model'' to the user, which the user can run locally on its device to select the item that best fits its private features. We study a setup where the server uses a deep neural network (trained on the Movielens 25M dataset as the ground truth for movie recommendation. In the multi-selection paradigm, the average recommendation utility is approximately 97\% of the optimal utility (as determined by the ground truth neural network) while maintaining a local differential privacy guarantee with $\epsilon$ ranging around 1 with respect to feature vectors of neighboring users. This is in comparison to an average recommendation utility of 91\% in the non-multi-selection regime under the same constraints.
- [23] arXiv:2504.07422 [pdf, other]
-
Title: The Role of Machine Learning in Reducing Healthcare Costs: The Impact of Medication Adherence and Preventive Care on Hospitalization ExpensesSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
This study reveals the important role of prevention care and medication adherence in reducing hospitalizations. By using a structured dataset of 1,171 patients, four machine learning models Logistic Regression, Gradient Boosting, Random Forest, and Artificial Neural Networks are applied to predict five-year hospitalization risk, with the Gradient Boosting model achieving the highest accuracy of 81.2%. The result demonstrated that patients with high medication adherence and consistent preventive care can reduce 38.3% and 37.7% in hospitalization risk. The finding also suggests that targeted preventive care can have positive Return on Investment (ROI), and therefore ML models can effectively direct personalized interventions and contribute to long-term medical savings.
- [24] arXiv:2504.07437 [pdf, html, other]
-
Title: Unifying and extending Diffusion Models through PDEs for solving Inverse ProblemsAgnimitra Dasgupta, Alexsander Marciano da Cunha, Ali Fardisi, Mehrnegar Aminy, Brianna Binder, Bryan Shaddy, Assad A OberaiSubjects: Machine Learning (cs.LG); Computation (stat.CO); Machine Learning (stat.ML)
Diffusion models have emerged as powerful generative tools with applications in computer vision and scientific machine learning (SciML), where they have been used to solve large-scale probabilistic inverse problems. Traditionally, these models have been derived using principles of variational inference, denoising, statistical signal processing, and stochastic differential equations. In contrast to the conventional presentation, in this study we derive diffusion models using ideas from linear partial differential equations and demonstrate that this approach has several benefits that include a constructive derivation of the forward and reverse processes, a unified derivation of multiple formulations and sampling strategies, and the discovery of a new class of models. We also apply the conditional version of these models to solving canonical conditional density estimation problems and challenging inverse problems. These problems help establish benchmarks for systematically quantifying the performance of different formulations and sampling strategies in this study, and for future studies. Finally, we identify and implement a mechanism through which a single diffusion model can be applied to measurements obtained from multiple measurement operators. Taken together, the contents of this manuscript provide a new understanding and several new directions in the application of diffusion models to solving physics-based inverse problems.
- [25] arXiv:2504.07448 [pdf, html, other]
-
Title: LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank AdaptationComments: 24 pages, 7 figures, 20 tablesSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
Low-Rank Adaptation (LoRA) has emerged as a popular parameter-efficient fine-tuning (PEFT) method for Large Language Models (LLMs), yet it still incurs notable overhead and suffers from parameter interference in multi-task scenarios. We propose LoRA with Reduced Interference (LoRI), a simple yet effective approach that freezes the projection matrices $A$ as random projections and sparsifies the matrices $B$ using task-specific masks. This design substantially reduces the number of trainable parameters while maintaining strong task performance. Moreover, LoRI minimizes cross-task interference in adapter merging by leveraging the orthogonality between adapter subspaces, and supports continual learning by using sparsity to mitigate catastrophic forgetting. Extensive experiments across natural language understanding, mathematical reasoning, code generation, and safety alignment tasks demonstrate that LoRI outperforms full fine-tuning and existing PEFT methods, while using up to 95% fewer trainable parameters than LoRA. In multi-task experiments, LoRI enables effective adapter merging and continual learning with reduced cross-task interference. Code is available at: this https URL
- [26] arXiv:2504.07465 [pdf, html, other]
-
Title: Multi-Modal Data Fusion for Moisture Content Prediction in Apple DryingComments: Accepted for publication in the Proceedings of the 53rd North American Manufacturing Research Conference (NAMRC 53), to appear in Manufacturing LettersSubjects: Machine Learning (cs.LG)
Fruit drying is widely used in food manufacturing to reduce product moisture, ensure product safety, and extend product shelf life. Accurately predicting final moisture content (MC) is critically needed for quality control of drying processes. State-of-the-art methods can build deterministic relationships between process parameters and MC, but cannot adequately account for inherent process variabilities that are ubiquitous in fruit drying. To address this gap, this paper presents a novel multi-modal data fusion framework to effectively fuse two modalities of data: tabular data (process parameters) and high-dimensional image data (images of dried apple slices) to enable accurate MC prediction. The proposed modeling architecture permits flexible adjustment of information portion from tabular and image data modalities. Experimental validation shows that the multi-modal approach improves predictive accuracy substantially compared to state-of-the-art methods. The proposed method reduces root-mean-squared errors by 19.3%, 24.2%, and 15.2% over tabular-only, image-only, and standard tabular-image fusion models, respectively. Furthermore, it is demonstrated that our method is robust in varied tabular-image ratios and capable of effectively capturing inherent small-scale process variabilities. The proposed framework is extensible to a variety of other drying technologies.
- [27] arXiv:2504.07471 [pdf, html, other]
-
Title: Traversal Learning Coordination For Lossless And Efficient Distributed LearningSubjects: Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC)
In this paper, we introduce Traversal Learning (TL), a novel approach designed to address the problem of decreased quality encountered in popular distributed learning (DL) paradigms such as Federated Learning (FL), Split Learning (SL), and SplitFed Learning (SFL). Traditional FL experiences from an accuracy drop during aggregation due to its averaging function, while SL and SFL face increased loss due to the independent gradient updates on each split network. TL adopts a unique strategy where the model traverses the nodes during forward propagation (FP) and performs backward propagation (BP) on the orchestrator, effectively implementing centralized learning (CL) principles within a distributed environment. The orchestrator is tasked with generating virtual batches and planning the sequential node visits of the model during FP, aligning them with the ordered index of the data within these batches. We conducted experiments on six datasets representing diverse characteristics across various domains. Our evaluation demonstrates that TL is on par with classic CL approaches in terms of accurate inference, thereby offering a viable and robust solution for DL tasks. TL outperformed other DL methods and improved accuracy by 7.85% for independent and identically distributed (IID) datasets, macro F1-score by 1.06% for non-IID datasets, accuracy by 2.60% for text classification, and AUC by 3.88% and 4.54% for medical and financial datasets, respectively. By effectively preserving data privacy while maintaining performance, TL represents a significant advancement in DL methodologies.
- [28] arXiv:2504.07494 [pdf, html, other]
-
Title: Apt-Serve: Adaptive Request Scheduling on Hybrid Cache for Scalable LLM Inference ServingSubjects: Machine Learning (cs.LG)
Large language model (LLM) inference serving systems are essential to various LLM-based applications. As demand for LLM services continues to grow, scaling these systems to handle high request rates while meeting latency Service-Level Objectives (SLOs), referred to as effective throughput, becomes critical. However, existing systems often struggle to improve effective throughput, primarily due to a significant decline in Time To First Token (TTFT) SLO attainment. We identify two major causes of this bottleneck: (1) memory-intensive KV cache that limits batch size expansion under GPU memory constraints, and (2) rigid batch composition enforced by the default First-Come-First-Serve scheduling policy. In this paper, we introduce Apt-Serve, a scalable framework designed to enhance effective throughput in LLM inference serving. Apt-Serve features a new hybrid cache scheme that combines KV cache with a memory-efficient hidden cache for reusable input hidden state vectors, allowing large batch sizes and improving request concurrency. Based on the hybrid cache, Apt-Serve employs an adaptive runtime scheduling mechanism that dynamically optimizes batch composition. We formally define the adaptive scheduling optimization problem and propose an efficient algorithm with theoretical guarantees. Extensive evaluations on three real-world datasets and LLMs ranging from 13B to 66B parameters demonstrate that Apt-Serve achieves up to 8.8x improvement in effective throughput compared to the state-of-the-art inference serving systems.
- [29] arXiv:2504.07513 [pdf, html, other]
-
Title: GPT Carry-On: Training Foundation Model for Customization Could Be Simple, Scalable and AffordableSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (stat.ML)
Modern large language foundation models (LLM) have now entered the daily lives of millions of users. We ask a natural question whether it is possible to customize LLM for every user or every task. From system and industrial economy consideration, general continue-training or fine-tuning still require substantial computation and memory of training GPU nodes, whereas most inference nodes under deployment, possibly with lower-end GPUs, are configured to make forward pass fastest possible. We propose a framework to take full advantages of existing LLMs and systems of online service. We train an additional branch of transformer blocks on the final-layer embedding of pretrained LLMs, which is the base, then a carry-on module merge the base models to compose a customized LLM. We can mix multiple layers, or multiple LLMs specialized in different domains such as chat, coding, math, to form a new mixture of LLM that best fit a new task. As the base model don't need to update parameters, we are able to outsource most computation of the training job on inference nodes, and only train a lightweight carry-on on training nodes, where we consume less than 1GB GPU memory to train a 100M carry-on layer on 30B LLM. We tested Qwen and DeepSeek opensourced models for continue-pretraining and got faster loss convergence. We use it to improve solving math questions with extremely small computation and model size, with 1000 data samples of chain-of-thoughts, and as small as 1 MB parameters of two layer layer carry-on, and the results are promising.
- [30] arXiv:2504.07522 [pdf, html, other]
-
Title: Adversarial Subspace Generation for Outlier Detection in High-Dimensional DataJose Cribeiro-Ramallo, Federico Matteucci, Paul Enciu, Alexander Jenke, Vadim Arzamasov, Thorsten Strufe, Klemens BöhmComments: 35 pages, pre-printSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Statistics Theory (math.ST)
Outlier detection in high-dimensional tabular data is challenging since data is often distributed across multiple lower-dimensional subspaces -- a phenomenon known as the Multiple Views effect (MV). This effect led to a large body of research focused on mining such subspaces, known as subspace selection. However, as the precise nature of the MV effect was not well understood, traditional methods had to rely on heuristic-driven search schemes that struggle to accurately capture the true structure of the data. Properly identifying these subspaces is critical for unsupervised tasks such as outlier detection or clustering, where misrepresenting the underlying data structure can hinder the performance. We introduce Myopic Subspace Theory (MST), a new theoretical framework that mathematically formulates the Multiple Views effect and writes subspace selection as a stochastic optimization problem. Based on MST, we introduce V-GAN, a generative method trained to solve such an optimization problem. This approach avoids any exhaustive search over the feature space while ensuring that the intrinsic data structure is preserved. Experiments on 42 real-world datasets show that using V-GAN subspaces to build ensemble methods leads to a significant increase in one-class classification performance -- compared to existing subspace selection, feature selection, and embedding methods. Further experiments on synthetic data show that V-GAN identifies subspaces more accurately while scaling better than other relevant subspace selection methods. These results confirm the theoretical guarantees of our approach and also highlight its practical viability in high-dimensional settings.
- [31] arXiv:2504.07540 [pdf, html, other]
-
Title: PoGO: A Scalable Proof of Useful Work via Quantized Gradient Descent and Merkle ProofsComments: 14 pages, 1 figure, 1 tableSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
We present a design called \emph{Proof of Gradient Optimization} (PoGO) for blockchain consensus, where miners produce verifiable evidence of training large-scale machine-learning models. Building on previous work, we incorporate \emph{quantized gradients} (4-bit precision) to reduce storage and computation requirements, while still preserving the ability of verifiers to check that real progress has been made on lowering the model's loss. Additionally, we employ Merkle proofs over the full 32-bit model to handle large parameter sets and to enable random leaf checks with minimal on-chain data. We illustrate these ideas using GPT-3 (175B parameters) as a reference example and also refer to smaller but high-performance models (e.g., \emph{Gemma~3} with 27B parameters). We provide an empirical cost analysis showing that verification is significantly cheaper than training, thanks in part to quantization and sampling. We also discuss the necessity of longer block times (potentially hours) when incorporating meaningful training steps, the trade-offs when using specialized GPU hardware, and how binary diffs may incrementally optimize updates. Finally, we note that fine-tuning can be handled in a similar manner, merely changing the dataset and the manner of sampling but preserving the overall verification flow. Our protocol allows verifiers to issue either \emph{positive} or \emph{negative} attestations; these are aggregated at finalization to either confirm the update or slash the miner.
- [32] arXiv:2504.07557 [pdf, html, other]
-
Title: Using LLMs for Analyzing AIS DataSubjects: Machine Learning (cs.LG)
Recent research in Large Language Models (LLMs), has had a profound impact across various fields, including mobility data science. This paper explores the and experiment with different approaches to using LLMs for analyzing AIS data. We propose a set of carefully designed queries to assess the reasoning capabilities of LLMs in this kind of tasks. Further, we experiment with four different methods: (1) using LLMs as a natural language interface to a spatial database, (2) reasoning on raw data, (3) reasoning on compressed trajectories, and (4) reasoning on semantic trajectories. We investigate the strengths and weaknesses for the four methods, and discuss the findings. The goal is to provide valuable insights for both researchers and practitioners on selecting the most appropriate LLM-based method depending on their specific data analysis objectives.
- [33] arXiv:2504.07566 [pdf, html, other]
-
Title: Diffusion Transformers for Tabular Data Time Series GenerationComments: 26 pages, 19 figures, 13 tablesSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Tabular data generation has recently attracted a growing interest due to its different application scenarios. However, generating time series of tabular data, where each element of the series depends on the others, remains a largely unexplored domain. This gap is probably due to the difficulty of jointly solving different problems, the main of which are the heterogeneity of tabular data (a problem common to non-time-dependent approaches) and the variable length of a time series. In this paper, we propose a Diffusion Transformers (DiTs) based approach for tabular data series generation. Inspired by the recent success of DiTs in image and video generation, we extend this framework to deal with heterogeneous data and variable-length sequences. Using extensive experiments on six datasets, we show that the proposed approach outperforms previous work by a large margin.
- [34] arXiv:2504.07611 [pdf, html, other]
-
Title: Conditional Conformal Risk AdaptationSubjects: Machine Learning (cs.LG)
Uncertainty quantification is becoming increasingly important in image segmentation, especially for high-stakes applications like medical imaging. While conformal risk control generalizes conformal prediction beyond standard miscoverage to handle various loss functions such as false negative rate, its application to segmentation often yields inadequate conditional risk control: some images experience very high false negative rates while others have negligibly small ones. We develop Conformal Risk Adaptation (CRA), which introduces a new score function for creating adaptive prediction sets that significantly improve conditional risk control for segmentation tasks. We establish a novel theoretical framework that demonstrates a fundamental connection between conformal risk control and conformal prediction through a weighted quantile approach, applicable to any score function. To address the challenge of poorly calibrated probabilities in segmentation models, we introduce a specialized probability calibration framework that enhances the reliability of pixel-wise inclusion estimates. Using these calibrated probabilities, we propose Calibrated Conformal Risk Adaptation (CCRA) and a stratified variant (CCRA-S) that partitions images based on their characteristics and applies group-specific thresholds to further enhance conditional risk control. Our experiments on polyp segmentation demonstrate that all three methods (CRA, CCRA, and CCRA-S) provide valid marginal risk control and deliver more consistent conditional risk control across diverse images compared to standard approaches, offering a principled approach to uncertainty quantification that is particularly valuable for high-stakes and personalized segmentation applications.
- [35] arXiv:2504.07618 [pdf, other]
-
Title: CTSR: Cartesian tensor-based sparse regression for data-driven discovery of high-dimensional invariant governing equationsSubjects: Machine Learning (cs.LG)
Accurate and concise governing equations are crucial for understanding system dynamics. Recently, data-driven methods such as sparse regression have been employed to automatically uncover governing equations from data, representing a significant shift from traditional first-principles modeling. However, most existing methods focus on scalar equations, limiting their applicability to simple, low-dimensional scenarios, and failing to ensure rotation and reflection invariance without incurring significant computational cost or requiring additional prior knowledge. This paper proposes a Cartesian tensor-based sparse regression (CTSR) technique to accurately and efficiently uncover complex, high-dimensional governing equations while ensuring invariance. Evaluations on two two-dimensional (2D) and two three-dimensional (3D) test cases demonstrate that the proposed method achieves superior accuracy and efficiency compared to the conventional technique.
- [36] arXiv:2504.07625 [pdf, html, other]
-
Title: Deep Learning Meets Teleconnections: Improving S2S Predictions for European Winter WeatherComments: 21 pages, 6 figuresSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Predictions on subseasonal-to-seasonal (S2S) timescales--ranging from two weeks to two month--are crucial for early warning systems but remain challenging owing to chaos in the climate system. Teleconnections, such as the stratospheric polar vortex (SPV) and Madden-Julian Oscillation (MJO), offer windows of enhanced predictability, however, their complex interactions remain underutilized in operational forecasting. Here, we developed and evaluated deep learning architectures to predict North Atlantic-European (NAE) weather regimes, systematically assessing the role of remote drivers in improving S2S forecast skill of deep learning models. We implemented (1) a Long Short-term Memory (LSTM) network predicting the NAE regimes of the next six weeks based on previous regimes, (2) an Index-LSTM incorporating SPV and MJO indices, and (3) a ViT-LSTM using a Vision Transformer to directly encode stratospheric wind and tropical outgoing longwave radiation fields. These models are compared with operational hindcasts as well as other AI models. Our results show that leveraging teleconnection information enhances skill at longer lead times. Notably, the ViT-LSTM outperforms ECMWF's subseasonal hindcasts beyond week 4 by improving Scandinavian Blocking (SB) and Atlantic Ridge (AR) predictions. Analysis of high-confidence predictions reveals that NAO-, SB, and AR opportunity forecasts can be associated with SPV variability and MJO phase patterns aligning with established pathways, also indicating new patterns. Overall, our work demonstrates that encoding physically meaningful climate fields can enhance S2S prediction skill, advancing AI-driven subseasonal forecast. Moreover, the experiments highlight the potential of deep learning methods as investigative tools, providing new insights into atmospheric dynamics and predictability.
- [37] arXiv:2504.07633 [pdf, html, other]
-
Title: Kernel Logistic Regression Learning for High-Capacity Hopfield NetworksComments: submitted to IEICE journalSubjects: Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)
Hebbian learning limits Hopfield network storage capacity (pattern-to-neuron ratio around 0.14). We propose Kernel Logistic Regression (KLR) learning. Unlike linear methods, KLR uses kernels to implicitly map patterns to high-dimensional feature space, enhancing separability. By learning dual variables, KLR dramatically improves storage capacity, achieving perfect recall even when pattern numbers exceed neuron numbers (up to ratio 1.5 shown), and enhances noise robustness. KLR demonstrably outperforms Hebbian and linear logistic regression approaches.
- [38] arXiv:2504.07638 [pdf, html, other]
-
Title: Predicting the Lifespan of Industrial Printheads with Survival AnalysisSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Accurately predicting the lifespan of critical device components is essential for maintenance planning and production optimization, making it a topic of significant interest in both academia and industry. In this work, we investigate the use of survival analysis for predicting the lifespan of production printheads developed by Canon Production Printing. Specifically, we focus on the application of five techniques to estimate survival probabilities and failure rates: the Kaplan-Meier estimator, Cox proportional hazard model, Weibull accelerated failure time model, random survival forest, and gradient boosting. The resulting estimates are further refined using isotonic regression and subsequently aggregated to determine the expected number of failures. The predictions are then validated against real-world ground truth data across multiple time windows to assess model reliability. Our quantitative evaluation using three performance metrics demonstrates that survival analysis outperforms industry-standard baseline methods for printhead lifespan prediction.
- [39] arXiv:2504.07645 [pdf, html, other]
-
Title: Prediction of Usage Probabilities of Shopping-Mall Corridors Using Heterogeneous Graph Neural NetworksComments: 17 pages, working manuscript with partial resultsSubjects: Machine Learning (cs.LG)
We present a method based on graph neural network (GNN) for prediction of probabilities of usage of shopping-mall corridors. The heterogeneous graph network of shops and corridor paths are obtained from floorplans of the malls by creating vector layers for corridors, shops and entrances. These are subsequently assimilated into nodes and edges of graphs. The prediction of the usage probability is based on the shop features, namely, the area and usage categories they fall into, and on the graph connecting these shops, corridor junctions and entrances by corridor paths. Though the presented method is applicable for training on datasets obtained from a field survey or from pedestrian-detecting sensors, the target data of the supervised deep-learning work flow in this work are obtained from a probability method. We also include a context-specific representation learning of latent features. The usage-probability prediction is made on each edge, which is a connection by a section of corridor path between the adjacent nodes representing the shops or corridor points. To create a feature for each edge, the hidden-layer feature vectors acquired in the message-passing GNN layers at the nodes of each edge are averaged and concatenated with the vector obtained by their multiplication. These edge-features are then passed to multilayer perceptrons (MLP) to make the final prediction of usage probability on each edge. The samples of synthetic learning dataset for each shopping mall are obtained by changing the shops' usage and area categories, and by subsequently feeding the graph into the probability model.
When including different shopping malls in a single dataset, we also propose to consider graph-level features to inform the model with specific identifying features of each mall. - [40] arXiv:2504.07654 [pdf, html, other]
-
Title: ms-Mamba: Multi-scale Mamba for Time-Series ForecastingSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
The problem of Time-series Forecasting is generally addressed by recurrent, Transformer-based and the recently proposed Mamba-based architectures. However, existing architectures generally process their input at a single temporal scale, which may be sub-optimal for many tasks where information changes over multiple time scales. In this paper, we introduce a novel architecture called Multi-scale Mamba (ms-Mamba) to address this gap. ms-Mamba incorporates multiple temporal scales by using multiple Mamba blocks with different sampling rates ($\Delta$s). Our experiments on many benchmarks demonstrate that ms-Mamba outperforms state-of-the-art approaches, including the recently proposed Transformer-based and Mamba-based models.
- [41] arXiv:2504.07691 [pdf, html, other]
-
Title: Distilling Knowledge from Heterogeneous Architectures for Semantic SegmentationComments: Accepted to AAAI 2025Subjects: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
Current knowledge distillation (KD) methods for semantic segmentation focus on guiding the student to imitate the teacher's knowledge within homogeneous architectures. However, these methods overlook the diverse knowledge contained in architectures with different inductive biases, which is crucial for enabling the student to acquire a more precise and comprehensive understanding of the data during distillation. To this end, we propose for the first time a generic knowledge distillation method for semantic segmentation from a heterogeneous perspective, named HeteroAKD. Due to the substantial disparities between heterogeneous architectures, such as CNN and Transformer, directly transferring cross-architecture knowledge presents significant challenges. To eliminate the influence of architecture-specific information, the intermediate features of both the teacher and student are skillfully projected into an aligned logits space. Furthermore, to utilize diverse knowledge from heterogeneous architectures and deliver customized knowledge required by the student, a teacher-student knowledge mixing mechanism (KMM) and a teacher-student knowledge evaluation mechanism (KEM) are introduced. These mechanisms are performed by assessing the reliability and its discrepancy between heterogeneous teacher-student knowledge. Extensive experiments conducted on three main-stream benchmarks using various teacher-student pairs demonstrate that our HeteroAKD outperforms state-of-the-art KD methods in facilitating distillation between heterogeneous architectures.
- [42] arXiv:2504.07711 [pdf, html, other]
-
Title: Merging Embedded Topics with Optimal Transport for Online Topic Modeling on Data StreamsComments: Paper under reviewSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Topic modeling is a key component in unsupervised learning, employed to identify topics within a corpus of textual data. The rapid growth of social media generates an ever-growing volume of textual data daily, making online topic modeling methods essential for managing these data streams that continuously arrive over time. This paper introduces a novel approach to online topic modeling named StreamETM. This approach builds on the Embedded Topic Model (ETM) to handle data streams by merging models learned on consecutive partial document batches using unbalanced optimal transport. Additionally, an online change point detection algorithm is employed to identify shifts in topics over time, enabling the identification of significant changes in the dynamics of text streams. Numerical experiments on simulated and real-world data show StreamETM outperforming competitors.
- [43] arXiv:2504.07719 [pdf, html, other]
-
Title: Counting Hours, Counting Losses: The Toll of Unpredictable Work Schedules on Financial SecuritySubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
Financial instability has become a significant issue in today's society. While research typically focuses on financial aspects, there is a tendency to overlook time-related aspects of unstable work schedules. The inability to rely on consistent work schedules leads to burnout, work-family conflicts, and financial shocks that directly impact workers' income and assets. Unforeseen fluctuations in earnings pose challenges in financial planning, affecting decisions on savings and spending and ultimately undermining individuals' long-term financial stability and well-being.
This issue is particularly evident in sectors where workers experience frequently changing schedules without sufficient notice, including those in the food service and retail sectors, part-time and hourly workers, and individuals with lower incomes. These groups are already more financially vulnerable, and the unpredictable nature of their schedules exacerbates their financial fragility.
Our objective is to understand how unforeseen fluctuations in earnings exacerbate financial fragility by investigating the extent to which individuals' financial management depends on their ability to anticipate and plan for the future. To address this question, we develop a simulation framework that models how individuals optimize utility amidst financial uncertainty and the imperative to avoid financial ruin. We employ online learning techniques, specifically adapting workers' consumption policies based on evolving information about their work schedules.
With this framework, we show both theoretically and empirically how a worker's capacity to anticipate schedule changes enhances their long-term utility. Conversely, the inability to predict future events can worsen workers' instability. Moreover, our framework enables us to explore interventions to mitigate the problem of schedule uncertainty and evaluate their effectiveness. - [44] arXiv:2504.07722 [pdf, html, other]
-
Title: Relaxing the Markov Requirements on Reinforcement Learning Under Weak Partial IgnorabilitySubjects: Machine Learning (cs.LG); Methodology (stat.ME)
Incomplete data, confounding effects, and violations of the Markov property are interrelated problems which are ubiquitous in Reinforcement Learning applications. We introduce the concept of ``partial ignorabilty" and leverage it to establish a novel convergence theorem for adaptive Reinforcement Learning. This theoretical result relaxes the Markov assumption on the stochastic process underlying conventional $Q$-learning, deploying a generalized form of the Robbins-Monro stochastic approximation theorem to establish optimality. This result has clear downstream implications for most active subfields of Reinforcement Learning, with clear paths for extension to the field of Causal Inference.
- [45] arXiv:2504.07793 [pdf, html, other]
-
Title: Revisiting Likelihood-Based Out-of-Distribution Detection by Modeling RepresentationsYifan Ding, Arturas Aleksandrauskas, Amirhossein Ahmadian, Jonas Unger, Fredrik Lindsten, Gabriel EilertsenSubjects: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
Out-of-distribution (OOD) detection is critical for ensuring the reliability of deep learning systems, particularly in safety-critical applications. Likelihood-based deep generative models have historically faced criticism for their unsatisfactory performance in OOD detection, often assigning higher likelihood to OOD data than in-distribution samples when applied to image data. In this work, we demonstrate that likelihood is not inherently flawed. Rather, several properties in the images space prohibit likelihood as a valid detection score. Given a sufficiently good likelihood estimator, specifically using the probability flow formulation of a diffusion model, we show that likelihood-based methods can still perform on par with state-of-the-art methods when applied in the representation space of pre-trained encoders. The code of our work can be found at $\href{this https URL}{\texttt{this https URL}}$.
- [46] arXiv:2504.07822 [pdf, html, other]
-
Title: DG-STMTL: A Novel Graph Convolutional Network for Multi-Task Spatio-Temporal Traffic ForecastingSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Spatio-temporal traffic prediction is crucial in intelligent transportation systems. The key challenge of accurate prediction is how to model the complex spatio-temporal dependencies and adapt to the inherent dynamics in data. Traditional Graph Convolutional Networks (GCNs) often struggle with static adjacency matrices that introduce domain bias or learnable matrices that may be overfitting to specific patterns. This challenge becomes more complex when considering Multi-Task Learning (MTL). While MTL has the potential to enhance prediction accuracy through task synergies, it can also face significant hurdles due to task interference. To overcome these challenges, this study introduces a novel MTL framework, Dynamic Group-wise Spatio-Temporal Multi-Task Learning (DG-STMTL). DG-STMTL proposes a hybrid adjacency matrix generation module that combines static matrices with dynamic ones through a task-specific gating mechanism. We also introduce a group-wise GCN module to enhance the modelling capability of spatio-temporal dependencies. We conduct extensive experiments on two real-world datasets to evaluate our method. Results show that our method outperforms other state-of-the-arts, indicating its effectiveness and robustness.
- [47] arXiv:2504.07835 [pdf, html, other]
-
Title: Pychop: Emulating Low-Precision Arithmetic in Numerical Methods and Neural NetworksSubjects: Machine Learning (cs.LG); Numerical Analysis (math.NA)
Motivated by the growing demand for low-precision arithmetic in computational science, we exploit lower-precision emulation in Python -- widely regarded as the dominant programming language for numerical analysis and machine learning. Low-precision training has revolutionized deep learning by enabling more efficient computation and reduced memory and energy consumption while maintaining model fidelity. To better enable numerical experimentation with and exploration of low precision computation, we developed the Pychop library, which supports customizable floating-point formats and a comprehensive set of rounding modes in Python, allowing users to benefit from fast, low-precision emulation in numerous applications. Pychop also introduces interfaces for both PyTorch and JAX, enabling efficient low-precision emulation on GPUs for neural network training and inference with unparalleled flexibility.
In this paper, we offer a comprehensive exposition of the design, implementation, validation, and practical application of Pychop, establishing it as a foundational tool for advancing efficient mixed-precision algorithms. Furthermore, we present empirical results on low-precision emulation for image classification and object detection using published datasets, illustrating the sensitivity of the use of low precision and offering valuable insights into its impact. Pychop enables in-depth investigations into the effects of numerical precision, facilitates the development of novel hardware accelerators, and integrates seamlessly into existing deep learning workflows. Software and experimental code are publicly available at this https URL. - [48] arXiv:2504.07863 [pdf, html, other]
-
Title: Robust Hallucination Detection in LLMs via Adaptive Token SelectionSubjects: Machine Learning (cs.LG)
Hallucinations in large language models (LLMs) pose significant safety concerns that impede their broader deployment. Recent research in hallucination detection has demonstrated that LLMs' internal representations contain truthfulness hints, which can be harnessed for detector training. However, the performance of these detectors is heavily dependent on the internal representations of predetermined tokens, fluctuating considerably when working on free-form generations with varying lengths and sparse distributions of hallucinated entities. To address this, we propose HaMI, a novel approach that enables robust detection of hallucinations through adaptive selection and learning of critical tokens that are most indicative of hallucinations. We achieve this robustness by an innovative formulation of the Hallucination detection task as Multiple Instance (HaMI) learning over token-level representations within a sequence, thereby facilitating a joint optimisation of token selection and hallucination detection on generation sequences of diverse forms. Comprehensive experimental results on four hallucination benchmarks show that HaMI significantly outperforms existing state-of-the-art approaches.
- [49] arXiv:2504.07891 [pdf, html, other]
-
Title: SpecReason: Fast and Accurate Inference-Time Compute via Speculative ReasoningSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Recent advances in inference-time compute have significantly improved performance on complex tasks by generating long chains of thought (CoTs) using Large Reasoning Models (LRMs). However, this improved accuracy comes at the cost of high inference latency due to the length of generated reasoning sequences and the autoregressive nature of decoding. Our key insight in tackling these overheads is that LRM inference, and the reasoning that it embeds, is highly tolerant of approximations: complex tasks are typically broken down into simpler steps, each of which brings utility based on the semantic insight it provides for downstream steps rather than the exact tokens it generates. Accordingly, we introduce SpecReason, a system that automatically accelerates LRM inference by using a lightweight model to (speculatively) carry out simpler intermediate reasoning steps and reserving the costly base model only to assess (and potentially correct) the speculated outputs. Importantly, SpecReason's focus on exploiting the semantic flexibility of thinking tokens in preserving final-answer accuracy is complementary to prior speculation techniques, most notably speculative decoding, which demands token-level equivalence at each step. Across a variety of reasoning benchmarks, SpecReason achieves 1.5-2.5$\times$ speedup over vanilla LRM inference while improving accuracy by 1.0-9.9\%. Compared to speculative decoding without SpecReason, their combination yields an additional 19.4-44.2\% latency reduction. We open-source SpecReason at this https URL.
- [50] arXiv:2504.07894 [pdf, html, other]
-
Title: DiverseFlow: Sample-Efficient Diverse Mode Coverage in FlowsSubjects: Machine Learning (cs.LG)
Many real-world applications of flow-based generative models desire a diverse set of samples that cover multiple modes of the target distribution. However, the predominant approach for obtaining diverse sets is not sample-efficient, as it involves independently obtaining many samples from the source distribution and mapping them through the flow until the desired mode coverage is achieved. As an alternative to repeated sampling, we introduce DiverseFlow: a training-free approach to improve the diversity of flow models. Our key idea is to employ a determinantal point process to induce a coupling between the samples that drives diversity under a fixed sampling budget. In essence, DiverseFlow allows exploration of more variations in a learned flow model with fewer samples. We demonstrate the efficacy of our method for tasks where sample-efficient diversity is desirable, such as text-guided image generation with polysemous words, inverse problems like large-hole inpainting, and class-conditional image synthesis.
- [51] arXiv:2504.07896 [pdf, html, other]
-
Title: Fast Adaptation with Behavioral Foundation ModelsHarshit Sikchi, Andrea Tirinzoni, Ahmed Touati, Yingchen Xu, Anssi Kanervisto, Scott Niekum, Amy Zhang, Alessandro Lazaric, Matteo PirottaComments: 25 pagesSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Robotics (cs.RO)
Unsupervised zero-shot reinforcement learning (RL) has emerged as a powerful paradigm for pretraining behavioral foundation models (BFMs), enabling agents to solve a wide range of downstream tasks specified via reward functions in a zero-shot fashion, i.e., without additional test-time learning or planning. This is achieved by learning self-supervised task embeddings alongside corresponding near-optimal behaviors and incorporating an inference procedure to directly retrieve the latent task embedding and associated policy for any given reward function. Despite promising results, zero-shot policies are often suboptimal due to errors induced by the unsupervised training process, the embedding, and the inference procedure. In this paper, we focus on devising fast adaptation strategies to improve the zero-shot performance of BFMs in a few steps of online interaction with the environment while avoiding any performance drop during the adaptation process. Notably, we demonstrate that existing BFMs learn a set of skills containing more performant policies than those identified by their inference procedure, making them well-suited for fast adaptation. Motivated by this observation, we propose both actor-critic and actor-only fast adaptation strategies that search in the low-dimensional task-embedding space of the pre-trained BFM to rapidly improve the performance of its zero-shot policies on any downstream task. Notably, our approach mitigates the initial "unlearning" phase commonly observed when fine-tuning pre-trained RL models. We evaluate our fast adaptation strategies on top of four state-of-the-art zero-shot RL methods in multiple navigation and locomotion domains. Our results show that they achieve 10-40% improvement over their zero-shot performance in a few tens of episodes, outperforming existing baselines.
- [52] arXiv:2504.07910 [pdf, html, other]
-
Title: Hodge Laplacians and Hodge Diffusion MapsComments: 53 Pages, comments are welcome!Subjects: Machine Learning (cs.LG)
We introduce Hodge Diffusion Maps, a novel manifold learning algorithm designed to analyze and extract topological information from high-dimensional data-sets. This method approximates the exterior derivative acting on differential forms, thereby providing an approximation of the Hodge Laplacian operator. Hodge Diffusion Maps extend existing non-linear dimensionality reduction techniques, including vector diffusion maps, as well as the theories behind diffusion maps and Laplacian Eigenmaps. Our approach captures higher-order topological features of the data-set by projecting it into lower-dimensional Euclidean spaces using the Hodge Laplacian. We develop a theoretical framework to estimate the approximation error of the exterior derivative, based on sample points distributed over a real manifold. Numerical experiments support and validate the proposed methodology.
- [53] arXiv:2504.07912 [pdf, html, other]
-
Title: Echo Chamber: RL Post-training Amplifies Behaviors Learned in PretrainingSubjects: Machine Learning (cs.LG)
Reinforcement learning (RL)-based fine-tuning has become a crucial step in post-training language models for advanced mathematical reasoning and coding. Following the success of frontier reasoning models, recent work has demonstrated that RL fine-tuning consistently improves performance, even in smaller-scale models; however, the underlying mechanisms driving these improvements are not well-understood. Understanding the effects of RL fine-tuning requires disentangling its interaction with pretraining data composition, hyperparameters, and model scale, but such problems are exacerbated by the lack of transparency regarding the training data used in many existing models. In this work, we present a systematic end-to-end study of RL fine-tuning for mathematical reasoning by training models entirely from scratch on different mixtures of fully open datasets. We investigate the effects of various RL fine-tuning algorithms (PPO, GRPO, and Expert Iteration) across models of different scales. Our study reveals that RL algorithms consistently converge towards a dominant output distribution, amplifying patterns in the pretraining data. We also find that models of different scales trained on the same data mixture will converge to distinct output distributions, suggesting that there are scale-dependent biases in model generalization. Moreover, we find that RL post-training on simpler questions can lead to performance gains on harder ones, indicating that certain reasoning capabilities generalize across tasks. Our findings show that small-scale proxies in controlled settings can elicit interesting insights regarding the role of RL in shaping language model behavior.
- [54] arXiv:2504.07916 [pdf, other]
-
Title: Semantically Encoding Activity Labels for Context-Aware Human Activity RecognitionComments: Percom 2025Subjects: Machine Learning (cs.LG)
Prior work has primarily formulated CA-HAR as a multi-label classification problem, where model inputs are time-series sensor data and target labels are binary encodings representing whether a given activity or context occurs. These CA-HAR methods either predicted each label independently or manually imposed relationships using graphs. However, both strategies often neglect an essential aspect: activity labels have rich semantic relationships. For instance, walking, jogging, and running activities share similar movement patterns but differ in pace and intensity, indicating that they are semantically related. Consequently, prior CA-HAR methods often struggled to accurately capture these inherent and nuanced relationships, particularly on datasets with noisy labels typically used for CA-HAR or situations where the ideal sensor type is unavailable (e.g., recognizing speech without audio sensors). To address this limitation, we propose SEAL, which leverage LMs to encode CA-HAR activity labels to capture semantic relationships. LMs generate vector embeddings that preserve rich semantic information from natural language. Our SEAL approach encodes input-time series sensor data from smart devices and their associated activity and context labels (text) as vector embeddings. During training, SEAL aligns the sensor data representations with their corresponding activity/context label embeddings in a shared embedding space. At inference time, SEAL performs a similarity search, returning the CA-HAR label with the embedding representation closest to the input data. Although LMs have been widely explored in other domains, surprisingly, their potential in CA-HAR has been underexplored, making our approach a novel contribution to the field. Our research opens up new possibilities for integrating more advanced LMs into CA-HAR tasks.
- [55] arXiv:2504.07952 [pdf, html, other]
-
Title: Dynamic Cheatsheet: Test-Time Learning with Adaptive MemoryComments: this https URLSubjects: Machine Learning (cs.LG); Computation and Language (cs.CL)
Despite their impressive performance on complex tasks, current language models (LMs) typically operate in a vacuum: Each input query is processed separately, without retaining insights from previous attempts. Here, we present Dynamic Cheatsheet (DC), a lightweight framework that endows a black-box LM with a persistent, evolving memory. Rather than repeatedly re-discovering or re-committing the same solutions and mistakes, DC enables models to store and reuse accumulated strategies, code snippets, and general problem-solving insights at inference time. This test-time learning enhances performance substantially across a range of tasks without needing explicit ground-truth labels or human feedback. Leveraging DC, Claude 3.5 Sonnet's accuracy more than doubled on AIME math exams once it began retaining algebraic insights across questions. Similarly, GPT-4o's success rate on Game of 24 increased from 10% to 99% after the model discovered and reused a Python-based solution. In tasks prone to arithmetic mistakes, such as balancing equations, DC enabled GPT-4o and Claude to reach near-perfect accuracy by recalling previously validated code, whereas their baselines stagnated around 50%. Beyond arithmetic challenges, DC yields notable accuracy gains on knowledge-demanding tasks. Claude achieved a 9% improvement in GPQA-Diamond and an 8% boost on MMLU-Pro problems. Crucially, DC's memory is self-curated, focusing on concise, transferable snippets rather than entire transcript. Unlike finetuning or static retrieval methods, DC adapts LMs' problem-solving skills on the fly, without modifying their underlying parameters. Overall, our findings present DC as a promising approach for augmenting LMs with persistent memory, bridging the divide between isolated inference events and the cumulative, experience-driven learning characteristic of human cognition.
- [56] arXiv:2504.07964 [pdf, html, other]
-
Title: C3PO: Critical-Layer, Core-Expert, Collaborative Pathway Optimization for Test-Time Expert Re-MixingSubjects: Machine Learning (cs.LG)
Mixture-of-Experts (MoE) Large Language Models (LLMs) suffer from severely sub-optimal expert pathways-our study reveals that naive expert selection learned from pretraining leaves a surprising 10-20% accuracy gap for improvement. Motivated by this observation, we develop a novel class of test-time optimization methods to re-weight or "re-mixing" the experts in different layers jointly for each test sample. Since the test sample's ground truth is unknown, we propose to optimize a surrogate objective defined by the sample's "successful neighbors" from a reference set of samples. We introduce three surrogates and algorithms based on mode-finding, kernel regression, and the average loss of similar reference samples/tasks. To reduce the cost of optimizing whole pathways, we apply our algorithms merely to the core experts' mixing weights in critical layers, which enjoy similar performance but save significant computation. This leads to "Critical-Layer, Core-Expert, Collaborative Pathway Optimization (C3PO)". We apply C3PO to two recent MoE LLMs and examine it on six widely-used benchmarks. It consistently improves the base model by 7-15% in accuracy and outperforms widely used test-time learning baselines, e.g., in-context learning and prompt/prefix tuning, by a large margin. Moreover, C3PO enables MoE LLMs with 1-3B active parameters to outperform LLMs of 7-9B parameters, hence improving MoE's advantages on efficiency. Our thorough ablation study further sheds novel insights on achieving test-time improvement on MoE.
- [57] arXiv:2504.07965 [pdf, html, other]
-
Title: Cat, Rat, Meow: On the Alignment of Language Model and Human Term-Similarity JudgmentsComments: ICLR 2025 Workshop on Representational Alignment (Re-Align)Subjects: Machine Learning (cs.LG); Computation and Language (cs.CL)
Small and mid-sized generative language models have gained increasing attention. Their size and availability make them amenable to being analyzed at a behavioral as well as a representational level, allowing investigations of how these levels interact. We evaluate 32 publicly available language models for their representational and behavioral alignment with human similarity judgments on a word triplet task. This provides a novel evaluation setting to probe semantic associations in language beyond common pairwise comparisons. We find that (1) even the representations of small language models can achieve human-level alignment, (2) instruction-tuned model variants can exhibit substantially increased agreement, (3) the pattern of alignment across layers is highly model dependent, and (4) alignment based on models' behavioral responses is highly dependent on model size, matching their representational alignment only for the largest evaluated models.
New submissions (showing 57 of 57 entries)
- [58] arXiv:2503.23943 (cross-list from cs.AR) [pdf, html, other]
-
Title: DOMAC: Differentiable Optimization for High-Speed Multipliers and Multiply-AccumulatorsChenhao Xue, Yi Ren, Jinwei Zhou, Kezhi Li, Chen Zhang, Yibo Lin, Lining Zhang, Qiang Xu, Guangyu SunComments: Accepted by ISEDA 2025Subjects: Hardware Architecture (cs.AR); Machine Learning (cs.LG)
Multipliers and multiply-accumulators (MACs) are fundamental building blocks for compute-intensive applications such as artificial intelligence. With the diminishing returns of Moore's Law, optimizing multiplier performance now necessitates process-aware architectural innovations rather than relying solely on technology scaling. In this paper, we introduce DOMAC, a novel approach that employs differentiable optimization for designing multipliers and MACs at specific technology nodes. DOMAC establishes an analogy between optimizing multi-staged parallel compressor trees and training deep neural networks. Building on this insight, DOMAC reformulates the discrete optimization challenge into a continuous problem by incorporating differentiable timing and area objectives. This formulation enables us to utilize existing deep learning toolkit for highly efficient implementation of the differentiable solver. Experimental results demonstrate that DOMAC achieves significant enhancements in both performance and area efficiency compared to state-of-the-art baselines and commercial IPs in multiplier and MAC designs.
- [59] arXiv:2504.07102 (cross-list from cs.IR) [pdf, html, other]
-
Title: Behavior Importance-Aware Graph Neural Architecture Search for Cross-Domain RecommendationChendi Ge, Xin Wang, Ziwei Zhang, Yijian Qin, Hong Chen, Haiyang Wu, Yang Zhang, Yuekui Yang, Wenwu ZhuComments: AAAI 2025 OralSubjects: Information Retrieval (cs.IR); Machine Learning (cs.LG)
Cross-domain recommendation (CDR) mitigates data sparsity and cold-start issues in recommendation systems. While recent CDR approaches using graph neural networks (GNNs) capture complex user-item interactions, they rely on manually designed architectures that are often suboptimal and labor-intensive. Additionally, extracting valuable behavioral information from source domains to improve target domain recommendations remains challenging. To address these challenges, we propose Behavior importance-aware Graph Neural Architecture Search (BiGNAS), a framework that jointly optimizes GNN architecture and data importance for CDR. BiGNAS introduces two key components: a Cross-Domain Customized Supernetwork and a Graph-Based Behavior Importance Perceptron. The supernetwork, as a one-shot, retrain-free module, automatically searches the optimal GNN architecture for each domain without the need for retraining. The perceptron uses auxiliary learning to dynamically assess the importance of source domain behaviors, thereby improving target domain recommendations. Extensive experiments on benchmark CDR datasets and a large-scale industry advertising dataset demonstrate that BiGNAS consistently outperforms state-of-the-art baselines. To the best of our knowledge, this is the first work to jointly optimize GNN architecture and behavior data importance for cross-domain recommendation.
- [60] arXiv:2504.07104 (cross-list from cs.IR) [pdf, html, other]
-
Title: Relevance Isn't All You Need: Scaling RAG Systems With Inference-Time Compute Via Multi-Criteria RerankingSubjects: Information Retrieval (cs.IR); Computation and Language (cs.CL); Machine Learning (cs.LG)
Modern Large Language Model (LLM) systems typically rely on Retrieval Augmented Generation (RAG) which aims to gather context that is useful for response generation. These RAG systems typically optimize strictly towards retrieving context that is maximally relevant to the query. However, conventional theory suggests that retrieval systems which seek to maximize context relevance without any additional explicit criteria can create information bottlenecks. We reaffirm this finding in the modern age of LLM's by showing that in standard RAG pipelines, maximizing for context relevance alone can degrade downstream response quality. In response, we show evaluations of existing RAG methods which account for both context relevance and answer quality. These evaluations introduce a novel finding that existing RAG systems scale poorly with inference time compute usage when considering our combined metric. We introduce "RErank BEyond reLevance (REBEL)", which enables RAG systems to scale with inference-time compute via injection of multi-criteria optimization using Chain-of-Thought prompting (and optionally Multi-Turn dialogue). Ultimately, this enables a new performance/speed tradeoff curve, where RAG systems are able to achieve both higher relevance of retrieved contexts and superior answer quality as inference time increases. Code for the implementation of our method in llama-index can be found at the following PR: this https URL. Code for running experiments using this llama-index implementation can be found at this https URL.
- [61] arXiv:2504.07107 (cross-list from cs.IR) [pdf, html, other]
-
Title: Guarding Digital Privacy: Exploring User Profiling and Security EnhancementsComments: 46 Pages, 8 tables, 9 figuresSubjects: Information Retrieval (cs.IR); Machine Learning (cs.LG)
User profiling, the practice of collecting user information for personalized recommendations, has become widespread, driving progress in technology. However, this growth poses a threat to user privacy, as devices often collect sensitive data without their owners' awareness. This article aims to consolidate knowledge on user profiling, exploring various approaches and associated challenges. Through the lens of two companies sharing user data and an analysis of 18 popular Android applications in India across various categories, including $\textit{Social, Education, Entertainment, Travel, Shopping and Others}$, the article unveils privacy vulnerabilities. Further, the article propose an enhanced machine learning framework, employing decision trees and neural networks, that improves state-of-the-art classifiers in detecting personal information exposure. Leveraging the XAI (explainable artificial intelligence) algorithm LIME (Local Interpretable Model-agnostic Explanations), it enhances interpretability, crucial for reliably identifying sensitive data. Results demonstrate a noteworthy performance boost, achieving a $75.01\%$ accuracy with a reduced training time of $3.62$ seconds for neural networks. Concluding, the paper suggests research directions to strengthen digital security measures.
- [62] arXiv:2504.07110 (cross-list from cs.IR) [pdf, html, other]
-
Title: DashCLIP: Leveraging multimodal models for generating semantic embeddings for DoorDashSubjects: Information Retrieval (cs.IR); Machine Learning (cs.LG)
Despite the success of vision-language models in various generative tasks, obtaining high-quality semantic representations for products and user intents is still challenging due to the inability of off-the-shelf models to capture nuanced relationships between the entities. In this paper, we introduce a joint training framework for product and user queries by aligning uni-modal and multi-modal encoders through contrastive learning on image-text data. Our novel approach trains a query encoder with an LLM-curated relevance dataset, eliminating the reliance on engagement history. These embeddings demonstrate strong generalization capabilities and improve performance across applications, including product categorization and relevance prediction. For personalized ads recommendation, a significant uplift in the click-through rate and conversion rate after the deployment further confirms the impact on key business metrics. We believe that the flexibility of our framework makes it a promising solution toward enriching the user experience across the e-commerce landscape.
- [63] arXiv:2504.07115 (cross-list from cs.CL) [pdf, html, other]
-
Title: EqualizeIR: Mitigating Linguistic Biases in Retrieval ModelsComments: NAACL 2025Journal-ref: NAACL 2025Subjects: Computation and Language (cs.CL); Machine Learning (cs.LG)
This study finds that existing information retrieval (IR) models show significant biases based on the linguistic complexity of input queries, performing well on linguistically simpler (or more complex) queries while underperforming on linguistically more complex (or simpler) queries. To address this issue, we propose EqualizeIR, a framework to mitigate linguistic biases in IR models. EqualizeIR uses a linguistically biased weak learner to capture linguistic biases in IR datasets and then trains a robust model by regularizing and refining its predictions using the biased weak learner. This approach effectively prevents the robust model from overfitting to specific linguistic patterns in data. We propose four approaches for developing linguistically-biased models. Extensive experiments on several datasets show that our method reduces performance disparities across linguistically simple and complex queries, while improving overall retrieval performance.
- [64] arXiv:2504.07132 (cross-list from cs.CR) [pdf, other]
-
Title: SolRPDS: A Dataset for Analyzing Rug Pulls in Solana Decentralized FinanceComments: Accepted paper to appear in the 15th ACM Conference on Data and Application Security and Privacy (CODASPY 2025)Subjects: Cryptography and Security (cs.CR); Computational Engineering, Finance, and Science (cs.CE); Machine Learning (cs.LG)
Rug pulls in Solana have caused significant damage to users interacting with Decentralized Finance (DeFi). A rug pull occurs when developers exploit users' trust and drain liquidity from token pools on Decentralized Exchanges (DEXs), leaving users with worthless tokens. Although rug pulls in Ethereum and Binance Smart Chain (BSC) have gained attention recently, analysis of rug pulls in Solana remains largely under-explored. In this paper, we introduce SolRPDS (Solana Rug Pull Dataset), the first public rug pull dataset derived from Solana's transactions. We examine approximately four years of DeFi data (2021-2024) that covers suspected and confirmed tokens exhibiting rug pull patterns. The dataset, derived from 3.69 billion transactions, consists of 62,895 suspicious liquidity pools. The data is annotated for inactivity states, which is a key indicator, and includes several detailed liquidity activities such as additions, removals, and last interaction as well as other attributes such as inactivity periods and withdrawn token amounts, to help identify suspicious behavior. Our preliminary analysis reveals clear distinctions between legitimate and fraudulent liquidity pools and we found that 22,195 tokens in the dataset exhibit rug pull patterns during the examined period. SolRPDS can support a wide range of future research on rug pulls including the development of data-driven and heuristic-based solutions for real-time rug pull detection and mitigation.
- [65] arXiv:2504.07133 (cross-list from stat.ML) [pdf, other]
-
Title: Can SGD Select Good Fishermen? Local Convergence under Self-Selection Biases and BeyondSubjects: Machine Learning (stat.ML); Data Structures and Algorithms (cs.DS); Machine Learning (cs.LG); Statistics Theory (math.ST)
We revisit the problem of estimating $k$ linear regressors with self-selection bias in $d$ dimensions with the maximum selection criterion, as introduced by Cherapanamjeri, Daskalakis, Ilyas, and Zampetakis [CDIZ23, STOC'23]. Our main result is a $\operatorname{poly}(d,k,1/\varepsilon) + {k}^{O(k)}$ time algorithm for this problem, which yields an improvement in the running time of the algorithms of [CDIZ23] and [GM24, arXiv]. We achieve this by providing the first local convergence algorithm for self-selection, thus resolving the main open question of [CDIZ23].
To obtain this algorithm, we reduce self-selection to a seemingly unrelated statistical problem called coarsening. Coarsening occurs when one does not observe the exact value of the sample but only some set (a subset of the sample space) that contains the exact value. Inference from coarse samples arises in various real-world applications due to rounding by humans and algorithms, limited precision of instruments, and lag in multi-agent systems.
Our reduction to coarsening is intuitive and relies on the geometry of the self-selection problem, which enables us to bypass the limitations of previous analytic approaches. To demonstrate its applicability, we provide a local convergence algorithm for linear regression under another self-selection criterion, which is related to second-price auction data. Further, we give the first polynomial time local convergence algorithm for coarse Gaussian mean estimation given samples generated from a convex partition. Previously, only a sample-efficient algorithm was known due to Fotakis, Kalavasis, Kontonis, and Tzamos [FKKT21, COLT'21]. - [66] arXiv:2504.07156 (cross-list from q-bio.BM) [pdf, html, other]
-
Title: PLM-eXplain: Divide and Conquer the Protein Embedding SpaceSubjects: Biomolecules (q-bio.BM); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Protein language models (PLMs) have revolutionised computational biology through their ability to generate powerful sequence representations for diverse prediction tasks. However, their black-box nature limits biological interpretation and translation to actionable insights. We present an explainable adapter layer - PLM-eXplain (PLM-X), that bridges this gap by factoring PLM embeddings into two components: an interpretable subspace based on established biochemical features, and a residual subspace that preserves the model's predictive power. Using embeddings from ESM2, our adapter incorporates well-established properties, including secondary structure and hydropathy while maintaining high performance. We demonstrate the effectiveness of our approach across three protein-level classification tasks: prediction of extracellular vesicle association, identification of transmembrane helices, and prediction of aggregation propensity. PLM-X enables biological interpretation of model decisions without sacrificing accuracy, offering a generalisable solution for enhancing PLM interpretability across various downstream applications. This work addresses a critical need in computational biology by providing a bridge between powerful deep learning models and actionable biological insights.
- [67] arXiv:2504.07157 (cross-list from cs.NE) [pdf, html, other]
-
Title: GAAPO: Genetic Algorithmic Applied to Prompt OptimizationComments: 26 pages, 9 figuresSubjects: Neural and Evolutionary Computing (cs.NE); Machine Learning (cs.LG)
Large Language Models (LLMs) have demonstrated remarkable capabilities across various tasks, with their performance heavily dependent on the quality of input prompts \cite{schulhoff2025promptsurvey} \cite{sahoo2025promptengineering}. While prompt engineering has proven effective, it typically relies on manual adjustments, making it time-consuming and potentially suboptimal. This paper introduces GAAPO (Genetic Algorithm Applied to Prompt Optimization), a novel hybrid optimization framework that leverages genetic algorithm \cite{dejong1988gen} principles to evolve prompts through successive generations. Unlike traditional genetic approaches that rely solely on mutation and crossover operations, GAAPO integrates multiple specialized prompt generation strategies within its evolutionary framework. Through extensive experimentation on diverse datasets including ETHOS, MMLU-Pro, and GPQA, our analysis reveals several important point for the future development of automatic prompt optimization methods: importance of the tradeoff between the population size and the number of generations, effect of selection methods on stability results, capacity of different LLMs and especially reasoning models to be able to automatically generate prompts from similar queries... Furthermore, we provide insights into the relative effectiveness of different prompt generation strategies and their evolution across optimization phases. These findings contribute to both the theoretical understanding of prompt optimization and practical applications in improving LLM performance.
- [68] arXiv:2504.07163 (cross-list from cs.MA) [pdf, html, other]
-
Title: Multi-Object Tracking for Collision Avoidance Using Multiple Cameras in Open RAN NetworksSubjects: Multiagent Systems (cs.MA); Machine Learning (cs.LG); Robotics (cs.RO)
This paper deals with the multi-object detection and tracking problem, within the scope of open Radio Access Network (RAN), for collision avoidance in vehicular scenarios. To this end, a set of distributed intelligent agents collocated with cameras are considered. The fusion of detected objects is done at an edge service, considering Open RAN connectivity. Then, the edge service predicts the objects trajectories for collision avoidance. Compared to the related work a more realistic Open RAN network is implemented and multiple cameras are used.
- [69] arXiv:2504.07164 (cross-list from cs.SE) [pdf, html, other]
-
Title: R2E-Gym: Procedural Environments and Hybrid Verifiers for Scaling Open-Weights SWE AgentsComments: Website: this https URLSubjects: Software Engineering (cs.SE); Computation and Language (cs.CL); Machine Learning (cs.LG)
Improving open-source models on real-world SWE tasks (solving GITHUB issues) faces two key challenges: 1) scalable curation of execution environments to train these models, and, 2) optimal scaling of test-time compute. We introduce AgentGym, the largest procedurally-curated executable gym environment for training real-world SWE-agents, consisting of more than 8.7K tasks. AgentGym is powered by two main contributions: 1) SYNGEN: a synthetic data curation recipe that enables scalable curation of executable environments using test-generation and back-translation directly from commits, thereby reducing reliance on human-written issues or unit tests. We show that this enables more scalable training leading to pass@1 performance of 34.4% on SWE-Bench Verified benchmark with our 32B model. 2) Hybrid Test-time Scaling: we provide an in-depth analysis of two test-time scaling axes; execution-based and execution-free verifiers, demonstrating that they exhibit complementary strengths and limitations. Test-based verifiers suffer from low distinguishability, while execution-free verifiers are biased and often rely on stylistic features. Surprisingly, we find that while each approach individually saturates around 42-43%, significantly higher gains can be obtained by leveraging their complementary strengths. Overall, our approach achieves 51% on the SWE-Bench Verified benchmark, reflecting a new state-of-the-art for open-weight SWE-agents and for the first time showing competitive performance with proprietary models such as o1, o1-preview and sonnet-3.5-v2 (with tools). We will open-source our environments, models, and agent trajectories.
- [70] arXiv:2504.07174 (cross-list from cs.CL) [pdf, html, other]
-
Title: HypoEval: Hypothesis-Guided Evaluation for Natural Language GenerationComments: 22 pages, 3 figures, code link: this https URLSubjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Large language models (LLMs) have demonstrated great potential for automating the evaluation of natural language generation. Previous frameworks of LLM-as-a-judge fall short in two ways: they either use zero-shot setting without consulting any human input, which leads to low alignment, or fine-tune LLMs on labeled data, which requires a non-trivial number of samples. Moreover, previous methods often provide little reasoning behind automated evaluations. In this paper, we propose HypoEval, Hypothesis-guided Evaluation framework, which first uses a small corpus of human evaluations to generate more detailed rubrics for human judgments and then incorporates a checklist-like approach to combine LLM's assigned scores on each decomposed dimension to acquire overall scores. With only 30 human evaluations, HypoEval achieves state-of-the-art performance in alignment with both human rankings (Spearman correlation) and human scores (Pearson correlation), on average outperforming G-Eval by 11.86% and fine-tuned Llama-3.1-8B-Instruct with at least 3 times more human evaluations by 11.95%. Furthermore, we conduct systematic studies to assess the robustness of HypoEval, highlighting its effectiveness as a reliable and interpretable automated evaluation framework.
- [71] arXiv:2504.07199 (cross-list from cs.CL) [pdf, html, other]
-
Title: SemEval-2025 Task 5: LLMs4Subjects -- LLM-based Automated Subject Tagging for a National Technical Library's Open-Access CatalogComments: 10 pages, 4 figures, Accepted as SemEval 2025 Task 5 description paperSubjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Digital Libraries (cs.DL); Machine Learning (cs.LG)
We present SemEval-2025 Task 5: LLMs4Subjects, a shared task on automated subject tagging for scientific and technical records in English and German using the GND taxonomy. Participants developed LLM-based systems to recommend top-k subjects, evaluated through quantitative metrics (precision, recall, F1-score) and qualitative assessments by subject specialists. Results highlight the effectiveness of LLM ensembles, synthetic data generation, and multilingual processing, offering insights into applying LLMs for digital library classification.
- [72] arXiv:2504.07210 (cross-list from cs.GR) [pdf, html, other]
-
Title: MESA: Text-Driven Terrain Generation Using Latent Diffusion and Global Copernicus DataPaul Borne--Pons (Adobe Research), Mikolaj Czerkawski (Asterisk Labs), Rosalie Martin (Adobe Research), Romain Rouffet (Adobe Research)Comments: Accepted at CVPR 2025 Workshop MORSESubjects: Graphics (cs.GR); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
Terrain modeling has traditionally relied on procedural techniques, which often require extensive domain expertise and handcrafted rules. In this paper, we present MESA - a novel data-centric alternative by training a diffusion model on global remote sensing data. This approach leverages large-scale geospatial information to generate high-quality terrain samples from text descriptions, showcasing a flexible and scalable solution for terrain generation. The model's capabilities are demonstrated through extensive experiments, highlighting its ability to generate realistic and diverse terrain landscapes. The dataset produced to support this work, the Major TOM Core-DEM extension dataset, is released openly as a comprehensive resource for global terrain data. The results suggest that data-driven models, trained on remote sensing data, can provide a powerful tool for realistic terrain modeling and generation.
- [73] arXiv:2504.07213 (cross-list from cs.NE) [pdf, html, other]
-
Title: Evolutionary algorithms meet self-supervised learning: a comprehensive surveySubjects: Neural and Evolutionary Computing (cs.NE); Machine Learning (cs.LG)
The number of studies that combine Evolutionary Machine Learning and self-supervised learning has been growing steadily in recent years. Evolutionary Machine Learning has been shown to help automate the design of machine learning algorithms and to lead to more reliable solutions. Self-supervised learning, on the other hand, has produced good results in learning useful features when labelled data is limited. This suggests that the combination of these two areas can help both in shaping evolutionary processes and in automating the design of deep neural networks, while also reducing the need for labelled data. Still, there are no detailed reviews that explain how Evolutionary Machine Learning and self-supervised learning can be used together. To help with this, we provide an overview of studies that bring these areas together. Based on this growing interest and the range of existing works, we suggest a new sub-area of research, which we call Evolutionary Self-Supervised Learning and introduce a taxonomy for it. Finally, we point out some of the main challenges and suggest directions for future research to help Evolutionary Self-Supervised Learning grow and mature as a field.
- [74] arXiv:2504.07221 (cross-list from nlin.CD) [pdf, html, other]
-
Title: Reservoir Computing with a Single Oscillating Gas Bubble: Emphasizing the Chaotic RegimeSubjects: Chaotic Dynamics (nlin.CD); Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE); Fluid Dynamics (physics.flu-dyn)
The rising computational and energy demands of artificial intelligence systems urge the exploration of alternative software and hardware solutions that exploit physical effects for computation. According to machine learning theory, a neural network-based computational system must exhibit nonlinearity to effectively model complex patterns and relationships. This requirement has driven extensive research into various nonlinear physical systems to enhance the performance of neural networks. In this paper, we propose and theoretically validate a reservoir computing system based on a single bubble trapped within a bulk of liquid. By applying an external acoustic pressure wave to both encode input information and excite the complex nonlinear dynamics, we showcase the ability of this single-bubble reservoir computing system to forecast complex benchmarking time series and undertake classification tasks with high accuracy. Specifically, we demonstrate that a chaotic physical regime of bubble oscillation proves to be the most effective for this kind of computations.
- [75] arXiv:2504.07235 (cross-list from astro-ph.EP) [pdf, other]
-
Title: Earth-like planet predictor: A machine learning approachComments: 11 pages, 5 figures, published in A&AJournal-ref: A&A, 696, A94 (2025)Subjects: Earth and Planetary Astrophysics (astro-ph.EP); Instrumentation and Methods for Astrophysics (astro-ph.IM); Machine Learning (cs.LG)
Searching for planets analogous to Earth in terms of mass and equilibrium temperature is currently the first step in the quest for habitable conditions outside our Solar System and, ultimately, the search for life in the universe. Future missions such as PLATO or LIFE will begin to detect and characterise these small, cold planets, dedicating significant observation time to them. The aim of this work is to predict which stars are most likely to host an Earth-like planet (ELP) to avoid blind searches, minimises detection times, and thus maximises the number of detections. Using a previous study on correlations between the presence of an ELP and the properties of its system, we trained a Random Forest to recognise and classify systems as 'hosting an ELP' or 'not hosting an ELP'. The Random Forest was trained and tested on populations of synthetic planetary systems derived from the Bern model, and then applied to real observed systems. The tests conducted on the machine learning (ML) model yield precision scores of up to 0.99, indicating that 99% of the systems identified by the model as having ELPs possess at least one. Among the few real observed systems that have been tested, 44 have been selected as having a high probability of hosting an ELP, and a quick study of the stability of these systems confirms that the presence of an Earth-like planet within them would leave them stable. The excellent results obtained from the tests conducted on the ML model demonstrate its ability to recognise the typical architectures of systems with or without ELPs within populations derived from the Bern model. If we assume that the Bern model adequately describes the architecture of real systems, then such a tool can prove indispensable in the search for Earth-like planets. A similar approach could be applied to other planetary system formation models to validate those predictions.
- [76] arXiv:2504.07245 (cross-list from cs.AI) [pdf, html, other]
-
Title: A new training approach for text classification in Mental Health: LatentGLossComments: 10 pages, 3 Figures, 4 TablesSubjects: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
This study presents a multi-stage approach to mental health classification by leveraging traditional machine learning algorithms, deep learning architectures, and transformer-based models. A novel data set was curated and utilized to evaluate the performance of various methods, starting with conventional classifiers and advancing through neural networks. To broaden the architectural scope, recurrent neural networks (RNNs) such as LSTM and GRU were also evaluated to explore their effectiveness in modeling sequential patterns in the data. Subsequently, transformer models such as BERT were fine-tuned to assess the impact of contextual embeddings in this domain. Beyond these baseline evaluations, the core contribution of this study lies in a novel training strategy involving a dual-model architecture composed of a teacher and a student network. Unlike standard distillation techniques, this method does not rely on soft label transfer; instead, it facilitates information flow through both the teacher model's output and its latent representations by modifying the loss function. The experimental results highlight the effectiveness of each modeling stage and demonstrate that the proposed loss function and teacher-student interaction significantly enhance the model's learning capacity in mental health prediction tasks.
- [77] arXiv:2504.07257 (cross-list from cs.AI) [pdf, html, other]
-
Title: Better Decisions through the Right Causal World ModelElisabeth Dillies, Quentin Delfosse, Jannis Blüml, Raban Emunds, Florian Peter Busch, Kristian KerstingComments: 5 pages including references, 2 figuresSubjects: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Reinforcement learning (RL) agents have shown remarkable performances in various environments, where they can discover effective policies directly from sensory inputs. However, these agents often exploit spurious correlations in the training data, resulting in brittle behaviours that fail to generalize to new or slightly modified environments. To address this, we introduce the Causal Object-centric Model Extraction Tool (COMET), a novel algorithm designed to learn the exact interpretable causal world models (CWMs). COMET first extracts object-centric state descriptions from observations and identifies the environment's internal states related to the depicted objects' properties. Using symbolic regression, it models object-centric transitions and derives causal relationships governing object dynamics. COMET further incorporates large language models (LLMs) for semantic inference, annotating causal variables to enhance interpretability.
By leveraging these capabilities, COMET constructs CWMs that align with the true causal structure of the environment, enabling agents to focus on task-relevant features. The extracted CWMs mitigate the danger of shortcuts, permitting the development of RL systems capable of better planning and decision-making across dynamic scenarios. Our results, validated in Atari environments such as Pong and Freeway, demonstrate the accuracy and robustness of COMET, highlighting its potential to bridge the gap between object-centric reasoning and causal inference in reinforcement learning. - [78] arXiv:2504.07273 (cross-list from quant-ph) [pdf, html, other]
-
Title: Evaluating Parameter-Based Training Performance of Neural Networks and Variational Quantum CircuitsComments: Accepted at ICCS 2025Subjects: Quantum Physics (quant-ph); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
In recent years, neural networks (NNs) have driven significant advances in machine learning. However, as tasks grow more complex, NNs often require large numbers of trainable parameters, which increases computational and energy demands. Variational quantum circuits (VQCs) offer a promising alternative: they leverage quantum mechanics to capture intricate relationships and typically need fewer parameters. In this work, we evaluate NNs and VQCs on simple supervised and reinforcement learning tasks, examining models with different parameter sizes. We simulate VQCs and execute selected parts of the training process on real quantum hardware to approximate actual training times. Our results show that VQCs can match NNs in performance while using significantly fewer parameters, despite longer training durations. As quantum technology and algorithms advance, and VQC architectures improve, we posit that VQCs could become advantageous for certain machine learning tasks.
- [79] arXiv:2504.07285 (cross-list from cs.HC) [pdf, other]
-
Title: A Scalable Approach to Clustering Embedding ProjectionsComments: 4 pages, 4 figuresSubjects: Human-Computer Interaction (cs.HC); Machine Learning (cs.LG)
Interactive visualization of embedding projections is a useful technique for understanding data and evaluating machine learning models. Labeling data within these visualizations is critical for interpretation, as labels provide an overview of the projection and guide user navigation. However, most methods for producing labels require clustering the points, which can be computationally expensive as the number of points grows. In this paper, we describe an efficient clustering approach using kernel density estimation in the projected 2D space instead of points. This algorithm can produce high-quality cluster regions from a 2D density map in a few hundred milliseconds, orders of magnitude faster than current approaches. We contribute the design of the algorithm, benchmarks, and applications that demonstrate the utility of the algorithm, including labeling and summarization.
- [80] arXiv:2504.07334 (cross-list from cs.CV) [pdf, html, other]
-
Title: Objaverse++: Curated 3D Object Dataset with Quality AnnotationsChendi Lin, Heshan Liu, Qunshu Lin, Zachary Bright, Shitao Tang, Yihui He, Minghao Liu, Ling Zhu, Cindy LeComments: 8 pages, 8 figures. Accepted to CVPR 2025 Workshop on Efficient Large Vision Models (April 2025)Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
This paper presents Objaverse++, a curated subset of Objaverse enhanced with detailed attribute annotations by human experts. Recent advances in 3D content generation have been driven by large-scale datasets such as Objaverse, which contains over 800,000 3D objects collected from the Internet. Although Objaverse represents the largest available 3D asset collection, its utility is limited by the predominance of low-quality models. To address this limitation, we manually annotate 10,000 3D objects with detailed attributes, including aesthetic quality scores, texture color classifications, multi-object composition flags, transparency characteristics, etc. Then, we trained a neural network capable of annotating the tags for the rest of the Objaverse dataset. Through experiments and a user study on generation results, we demonstrate that models pre-trained on our quality-focused subset achieve better performance than those trained on the larger dataset of Objaverse in image-to-3D generation tasks. In addition, by comparing multiple subsets of training data filtered by our tags, our results show that the higher the data quality, the faster the training loss converges. These findings suggest that careful curation and rich annotation can compensate for the raw dataset size, potentially offering a more efficient path to develop 3D generative models. We release our enhanced dataset of approximately 500,000 curated 3D models to facilitate further research on various downstream tasks in 3D computer vision. In the near future, we aim to extend our annotations to cover the entire Objaverse dataset.
- [81] arXiv:2504.07341 (cross-list from quant-ph) [pdf, html, other]
-
Title: Learning to erase quantum states: thermodynamic implications of quantum learning theoryComments: 5.5 pages + 1 figureSubjects: Quantum Physics (quant-ph); Statistical Mechanics (cond-mat.stat-mech); Computational Complexity (cs.CC); Information Theory (cs.IT); Machine Learning (cs.LG)
The energy cost of erasing quantum states depends on our knowledge of the states. We show that learning algorithms can acquire such knowledge to erase many copies of an unknown state at the optimal energy cost. This is proved by showing that learning can be made fully reversible and has no fundamental energy cost itself. With simple counting arguments, we relate the energy cost of erasing quantum states to their complexity, entanglement, and magic. We further show that the constructed erasure protocol is computationally efficient when learning is efficient. Conversely, under standard cryptographic assumptions, we prove that the optimal energy cost cannot be achieved efficiently in general. These results also enable efficient work extraction based on learning. Together, our results establish a concrete connection between quantum learning theory and thermodynamics, highlighting the physical significance of learning processes and enabling efficient learning-based protocols for thermodynamic tasks.
- [82] arXiv:2504.07347 (cross-list from stat.ML) [pdf, html, other]
-
Title: Throughput-Optimal Scheduling Algorithms for LLM Inference and AI AgentsSubjects: Machine Learning (stat.ML); Machine Learning (cs.LG); Probability (math.PR)
As demand for Large Language Models (LLMs) and AI agents rapidly grows, optimizing systems for efficient LLM inference becomes critical. While significant efforts have targeted system-level engineering, little is explored through a mathematical modeling and queuing perspective.
In this paper, we aim to develop the queuing fundamentals for LLM inference, bridging the gap between queuing and LLM system communities. In particular, we study the throughput aspect in LLM inference systems. We prove that a large class of 'work-conserving' scheduling algorithms can achieve maximum throughput for both individual requests and AI agent workloads, highlighting 'work-conserving' as a key design principle in practice. Evaluations of real-world systems show that Orca and Sarathi-serve are throughput-optimal, reassuring practitioners, while FastTransformer and vanilla vLLM are not maximally stable and should be used with caution.
Our results highlight the substantial benefits queuing community can offer in improving LLM inference systems and call for more interdisciplinary developments. - [83] arXiv:2504.07360 (cross-list from cs.CL) [pdf, html, other]
-
Title: Enhancing Time Series Forecasting via Multi-Level Text Alignment with LLMsSubjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
The adaptation of large language models (LLMs) to time series forecasting poses unique challenges, as time series data is continuous in nature, while LLMs operate on discrete tokens. Despite the success of LLMs in natural language processing (NLP) and other structured domains, aligning time series data with language-based representations while maintaining both predictive accuracy and interpretability remains a significant hurdle. Existing methods have attempted to reprogram time series data into text-based forms, but these often fall short in delivering meaningful, interpretable results. In this paper, we propose a multi-level text alignment framework for time series forecasting using LLMs that not only improves prediction accuracy but also enhances the interpretability of time series representations. Our method decomposes time series into trend, seasonal, and residual components, which are then reprogrammed into component-specific text representations. We introduce a multi-level alignment mechanism, where component-specific embeddings are aligned with pre-trained word tokens, enabling more interpretable forecasts. Experiments on multiple datasets demonstrate that our method outperforms state-of-the-art models in accuracy while providing good interpretability.
- [84] arXiv:2504.07388 (cross-list from math.OC) [pdf, html, other]
-
Title: Min-Max Optimisation for Nonconvex-Nonconcave Functions Using a Random Zeroth-Order Extragradient AlgorithmSubjects: Optimization and Control (math.OC); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Numerical Analysis (math.NA)
This study explores the performance of the random Gaussian smoothing Zeroth-Order ExtraGradient (ZO-EG) scheme considering min-max optimisation problems with possibly NonConvex-NonConcave (NC-NC) objective functions. We consider both unconstrained and constrained, differentiable and non-differentiable settings. We discuss the min-max problem from the point of view of variational inequalities. For the unconstrained problem, we establish the convergence of the ZO-EG algorithm to the neighbourhood of an $\epsilon$-stationary point of the NC-NC objective function, whose radius can be controlled under a variance reduction scheme, along with its complexity. For the constrained problem, we introduce the new notion of proximal variational inequalities and give examples of functions satisfying this property. Moreover, we prove analogous results to the unconstrained case for the constrained problem. For the non-differentiable case, we prove the convergence of the ZO-EG algorithm to a neighbourhood of an $\epsilon$-stationary point of the smoothed version of the objective function, where the radius of the neighbourhood can be controlled, which can be related to the ($\delta,\epsilon$)-Goldstein stationary point of the original objective function.
- [85] arXiv:2504.07415 (cross-list from cs.CV) [pdf, other]
-
Title: Leveraging LLMs for Multimodal Retrieval-Augmented Radiology Report Generation via Key Phrase ExtractionSubjects: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL); Machine Learning (cs.LG)
Automated radiology report generation (RRG) holds potential to reduce radiologists' workload, especially as recent advancements in large language models (LLMs) enable the development of multimodal models for chest X-ray (CXR) report generation. However, multimodal LLMs (MLLMs) are resource-intensive, requiring vast datasets and substantial computational cost for training. To address these challenges, we propose a retrieval-augmented generation approach that leverages multimodal retrieval and LLMs to generate radiology reports while mitigating hallucinations and reducing computational demands. Our method uses LLMs to extract key phrases from radiology reports, effectively focusing on essential diagnostic information. Through exploring effective training strategies, including image encoder structure search, adding noise to text embeddings, and additional training objectives, we combine complementary pre-trained image encoders and adopt contrastive learning between text and semantic image embeddings. We evaluate our approach on MIMIC-CXR dataset, achieving state-of-the-art results on CheXbert metrics and competitive RadGraph F1 metric alongside MLLMs, without requiring LLM fine-tuning. Our method demonstrates robust generalization for multi-view RRG, making it suitable for comprehensive clinical applications.
- [86] arXiv:2504.07416 (cross-list from cs.CV) [pdf, html, other]
-
Title: RadZero: Similarity-Based Cross-Attention for Explainable Vision-Language Alignment in Radiology with Zero-Shot Multi-Task CapabilitySubjects: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL); Machine Learning (cs.LG)
Recent advancements in multi-modal models have significantly improved vision-language alignment in radiology. However, existing approaches struggle to effectively utilize complex radiology reports for learning, rely on low-resolution images, and offer limited interpretability in attention mechanisms. To address these challenges, we introduce RadZero, a novel similarity-based cross-attention framework for vision-language alignment in radiology with zero-shot multi-task capability. RadZero leverages large language models to extract minimal semantic sentences from radiology reports and employs a multi-positive contrastive learning strategy to effectively capture relationships between images and multiple relevant textual descriptions. It also utilizes a pre-trained vision encoder with additional trainable Transformer layers, allowing efficient high-resolution image processing. By computing similarity between text embeddings and local image patch features, RadZero enables zero-shot inference with similarity probability for classification and pixel-level cross-modal similarity maps for grounding and segmentation. Experimental results on public chest radiograph benchmarks show that RadZero outperforms state-of-the-art methods in zero-shot classification, grounding, and segmentation. Furthermore, cross-modal similarity map analysis highlights its potential for improving explainability in vision-language alignment. Additionally, qualitative evaluation demonstrates RadZero's capability for open-vocabulary semantic segmentation, further validating its effectiveness in medical imaging.
- [87] arXiv:2504.07425 (cross-list from cs.AI) [pdf, other]
-
Title: Enhancing Player Enjoyment with a Two-Tier DRL and LLM-Based Agent System for Fighting GamesComments: 15 pages, 8 figures. Submitted to a peer-reviewed conference, under reviewSubjects: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Deep reinforcement learning (DRL) has effectively enhanced gameplay experiences and game design across various game genres. However, few studies on fighting game agents have focused explicitly on enhancing player enjoyment, a critical factor for both developers and players. To address this gap and establish a practical baseline for designing enjoyability-focused agents, we propose a two-tier agent (TTA) system and conducted experiments in the classic fighting game Street Fighter II. The first tier of TTA employs a task-oriented network architecture, modularized reward functions, and hybrid training to produce diverse and skilled DRL agents. In the second tier of TTA, a Large Language Model Hyper-Agent, leveraging players' playing data and feedback, dynamically selects suitable DRL opponents. In addition, we investigate and model several key factors that affect the enjoyability of the opponent. The experiments demonstrate improvements from 64. 36% to 156. 36% in the execution of advanced skills over baseline methods. The trained agents also exhibit distinct game-playing styles. Additionally, we conducted a small-scale user study, and the overall enjoyment in the player's feedback validates the effectiveness of our TTA system.
- [88] arXiv:2504.07426 (cross-list from stat.ME) [pdf, html, other]
-
Title: Conditional Data Synthesis AugmentationSubjects: Methodology (stat.ME); Machine Learning (cs.LG)
Reliable machine learning and statistical analysis rely on diverse, well-distributed training data. However, real-world datasets are often limited in size and exhibit underrepresentation across key subpopulations, leading to biased predictions and reduced performance, particularly in supervised tasks such as classification. To address these challenges, we propose Conditional Data Synthesis Augmentation (CoDSA), a novel framework that leverages generative models, such as diffusion models, to synthesize high-fidelity data for improving model performance across multimodal domains including tabular, textual, and image data. CoDSA generates synthetic samples that faithfully capture the conditional distributions of the original data, with a focus on under-sampled or high-interest regions. Through transfer learning, CoDSA fine-tunes pre-trained generative models to enhance the realism of synthetic data and increase sample density in sparse areas. This process preserves inter-modal relationships, mitigates data imbalance, improves domain adaptation, and boosts generalization. We also introduce a theoretical framework that quantifies the statistical accuracy improvements enabled by CoDSA as a function of synthetic sample volume and targeted region allocation, providing formal guarantees of its effectiveness. Extensive experiments demonstrate that CoDSA consistently outperforms non-adaptive augmentation strategies and state-of-the-art baselines in both supervised and unsupervised settings.
- [89] arXiv:2504.07478 (cross-list from cs.CR) [pdf, other]
-
Title: Intelligent DoS and DDoS Detection: A Hybrid GRU-NTM Approach to Network SecurityCaroline Panggabean, Chandrasekar Venkatachalam, Priyanka Shah, Sincy John, Renuka Devi P, Shanmugavalli VenkatachalamComments: Accepted at the 2024 5th International Conference on Smart Electronics and Communication (ICOSEC). This is the accepted manuscript version. The final version is published by IEEE at this https URLSubjects: Cryptography and Security (cs.CR); Machine Learning (cs.LG)
Detecting Denial of Service (DoS) and Distributed Denial of Service (DDoS) attacks remains a critical challenge in cybersecurity. This research introduces a hybrid deep learning model combining Gated Recurrent Units (GRUs) and a Neural Turing Machine (NTM) for enhanced intrusion detection. Trained on the UNSW-NB15 and BoT-IoT datasets, the model employs GRU layers for sequential data processing and an NTM for long-term pattern recognition. The proposed approach achieves 99% accuracy in distinguishing between normal, DoS, and DDoS traffic. These findings offer promising advancements in real-time threat detection and contribute to improved network security across various domains.
- [90] arXiv:2504.07481 (cross-list from physics.ao-ph) [pdf, other]
-
Title: A Mechanism-Learning Deeply Coupled Model for Remote Sensing Retrieval of Global Land Surface TemperatureTian Xie, Menghui Jiang, Huanfeng Shen, Huifang Li, Cao Zeng, Xiaobin Guan, Jun Ma, Guanhao Zhang, Liangpei ZhangSubjects: Atmospheric and Oceanic Physics (physics.ao-ph); Machine Learning (cs.LG)
Land surface temperature (LST) retrieval from remote sensing data is pivotal for analyzing climate processes and surface energy budgets. However, LST retrieval is an ill-posed inverse problem, which becomes particularly severe when only a single band is available. In this paper, we propose a deeply coupled framework integrating mechanistic modeling and machine learning to enhance the accuracy and generalizability of single-channel LST retrieval. Training samples are generated using a physically-based radiative transfer model and a global collection of 5810 atmospheric profiles. A physics-informed machine learning framework is proposed to systematically incorporate the first principles from classical physical inversion models into the learning workflow, with optimization constrained by radiative transfer equations. Global validation demonstrated a 30% reduction in root-mean-square error versus standalone methods. Under extreme humidity, the mean absolute error decreased from 4.87 K to 2.29 K (53% improvement). Continental-scale tests across five continents confirmed the superior generalizability of this model.
- [91] arXiv:2504.07532 (cross-list from cs.CL) [pdf, html, other]
-
Title: AI-Slop to AI-Polish? Aligning Language Models through Edit-Based Writing Rewards and Test-time ComputationComments: Under SubmissionSubjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
AI-generated text is proliferating across domains, from creative writing and journalism to marketing content and scientific articles. Models can follow user-provided instructions to generate coherent and grammatically correct outputs but in this work, we study a more fundamental question: how do we evaluate and improve the writing quality of AI-generated text? Writing quality assessment has received less attention from the community, in part because it is fundamentally subjective and requires expertise. We first introduce the Writing Quality Benchmark (WQ) by consolidating five writing-preference datasets into 4,729 writing quality judgments. Our experiments show that competitive baselines, including state-of-the-art LLMs that excel at reasoning tasks, barely outperform random baselines on WQ. We then train specialized Writing Quality Reward Models (WQRM) of various sizes for writing quality assessment that demonstrate strong generalization on four out-of-distribution test sets and 74% accuracy on the WQ benchmark. To further show WQRM's practical benefits during inference, we leverage additional test-time compute to generate and rank multiple candidate revisions, allowing us to select higher-quality outputs from an initial draft. Human evaluation with 9 experienced writers confirm that WQRM-based selection produces writing samples preferred by experts 66% overall, and 72.2% when the reward gap is larger than 1 point. We release our datasets and models to encourage community engagement with writing quality assessment and development of AI writing systems better aligned with human preferences.
- [92] arXiv:2504.07560 (cross-list from eess.IV) [pdf, html, other]
-
Title: PhaseGen: A Diffusion-Based Approach for Complex-Valued MRI Data GenerationMoritz Rempe, Fabian Hörst, Helmut Becker, Marco Schlimbach, Lukas Rotkopf, Kevin Kröninger, Jens KleesiekSubjects: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
Magnetic resonance imaging (MRI) raw data, or k-Space data, is complex-valued, containing both magnitude and phase information. However, clinical and existing Artificial Intelligence (AI)-based methods focus only on magnitude images, discarding the phase data despite its potential for downstream tasks, such as tumor segmentation and classification. In this work, we introduce $\textit{PhaseGen}$, a novel complex-valued diffusion model for generating synthetic MRI raw data conditioned on magnitude images, commonly used in clinical practice. This enables the creation of artificial complex-valued raw data, allowing pretraining for models that require k-Space information. We evaluate PhaseGen on two tasks: skull-stripping directly in k-Space and MRI reconstruction using the publicly available FastMRI dataset. Our results show that training with synthetic phase data significantly improves generalization for skull-stripping on real-world data, with an increased segmentation accuracy from $41.1\%$ to $80.1\%$, and enhances MRI reconstruction when combined with limited real-world data. This work presents a step forward in utilizing generative AI to bridge the gap between magnitude-based datasets and the complex-valued nature of MRI raw data. This approach allows researchers to leverage the vast amount of avaliable image domain data in combination with the information-rich k-Space data for more accurate and efficient diagnostic tasks. We make our code publicly $\href{this https URL}{\text{available here}}$.
- [93] arXiv:2504.07567 (cross-list from cs.CV) [pdf, html, other]
-
Title: Benchmarking Image Embeddings for E-Commerce: Evaluating Off-the Shelf Foundation Models, Fine-Tuning Strategies and Practical Trade-offsComments: accepted at Future Technologies Conference (FTC 2025)Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE); Information Retrieval (cs.IR); Machine Learning (cs.LG)
We benchmark foundation models image embeddings for classification and retrieval in e-Commerce, evaluating their suitability for real-world applications. Our study spans embeddings from pre-trained convolutional and transformer models trained via supervised, self-supervised, and text-image contrastive learning. We assess full fine-tuning and transfer learning (top-tuning) on six diverse e-Commerce datasets: fashion, consumer goods, cars, food, and retail. Results show full fine-tuning consistently performs well, while text-image and self-supervised embeddings can match its performance with less training. While supervised embeddings remain stable across architectures, SSL and contrastive embeddings vary significantly, often benefiting from top-tuning. Top-tuning emerges as an efficient alternative to full fine-tuning, reducing computational costs. We also explore cross-tuning, noting its impact depends on dataset characteristics. Our findings offer practical guidelines for embedding selection and fine-tuning strategies, balancing efficiency and performance.
- [94] arXiv:2504.07578 (cross-list from cs.CR) [pdf, html, other]
-
Title: Privacy-Preserving Vertical K-Means ClusteringFederico Mazzone, Trevor Brown, Florian Kerschbaum, Kevin H. Wilson, Maarten Everts, Florian Hahn, Andreas PeterSubjects: Cryptography and Security (cs.CR); Machine Learning (cs.LG)
Clustering is a fundamental data processing task used for grouping records based on one or more features. In the vertically partitioned setting, data is distributed among entities, with each holding only a subset of those features. A key challenge in this scenario is that computing distances between records requires access to all distributed features, which may be privacy-sensitive and cannot be directly shared with other parties. The goal is to compute the joint clusters while preserving the privacy of each entity's dataset. Existing solutions using secret sharing or garbled circuits implement privacy-preserving variants of Lloyd's algorithm but incur high communication costs, scaling as O(nkt), where n is the number of data points, k the number of clusters, and t the number of rounds. These methods become impractical for large datasets or several parties, limiting their use to LAN settings only. On the other hand, a different line of solutions rely on differential privacy (DP) to outsource the local features of the parties to a central server. However, they often significantly degrade the utility of the clustering outcome due to excessive noise. In this work, we propose a novel solution based on homomorphic encryption and DP, reducing communication complexity to O(n+kt). In our method, parties securely outsource their features once, allowing a computing party to perform clustering operations under encryption. DP is applied only to the clusters' centroids, ensuring privacy with minimal impact on utility. Our solution clusters 100,000 two-dimensional points into five clusters using only 73MB of communication, compared to 101GB for existing works, and completes in just under 3 minutes on a 100Mbps network, whereas existing works take over 1 day. This makes our solution practical even for WAN deployments, all while maintaining accuracy comparable to plaintext k-means algorithms.
- [95] arXiv:2504.07583 (cross-list from cs.CL) [pdf, html, other]
-
Title: Do LLMs Understand Your Translations? Evaluating Paragraph-level MT with Question AnsweringSubjects: Computation and Language (cs.CL); Machine Learning (cs.LG)
Despite the steady progress in machine translation evaluation, existing automatic metrics struggle to capture how well meaning is preserved beyond sentence boundaries. We posit that reliance on a single intrinsic quality score, trained to mimic human judgments, might be insufficient for evaluating translations of long, complex passages, and a more ``pragmatic'' approach that assesses how accurately key information is conveyed by a translation in context is needed. We introduce TREQA (Translation Evaluation via Question-Answering), a framework that extrinsically evaluates translation quality by assessing how accurately candidate translations answer reading comprehension questions that target key information in the original source or reference texts. In challenging domains that require long-range understanding, such as literary texts, we show that TREQA is competitive with and, in some cases, outperforms state-of-the-art neural and LLM-based metrics in ranking alternative paragraph-level translations, despite never being explicitly optimized to correlate with human judgments. Furthermore, the generated questions and answers offer interpretability: empirical analysis shows that they effectively target translation errors identified by experts in evaluated datasets. Our code is available at this https URL
- [96] arXiv:2504.07607 (cross-list from math.OC) [pdf, html, other]
-
Title: Stochastic Smoothed Primal-Dual Algorithms for Nonconvex Optimization with Linear Inequality ConstraintsSubjects: Optimization and Control (math.OC); Machine Learning (cs.LG)
We propose smoothed primal-dual algorithms for solving stochastic and smooth nonconvex optimization problems with linear inequality constraints. Our algorithms are single-loop and only require a single stochastic gradient based on one sample at each iteration. A distinguishing feature of our algorithm is that it is based on an inexact gradient descent framework for the Moreau envelope, where the gradient of the Moreau envelope is estimated using one step of a stochastic primal-dual augmented Lagrangian method. To handle inequality constraints and stochasticity, we combine the recently established global error bounds in constrained optimization with a Moreau envelope-based analysis of stochastic proximal algorithms. For obtaining $\varepsilon$-stationary points, we establish the optimal $O(\varepsilon^{-4})$ sample complexity guarantee for our algorithms and provide extensions to stochastic linear constraints. We also show how to improve this complexity to $O(\varepsilon^{-3})$ by using variance reduction and the expected smoothness assumption. Unlike existing methods, the iterations of our algorithms are free of subproblems, large batch sizes or increasing penalty parameters and use dual variable updates to ensure feasibility.
- [97] arXiv:2504.07619 (cross-list from cs.AI) [pdf, html, other]
-
Title: Beating Transformers using Synthetic CognitionSubjects: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
The road to Artificial General Intelligence goes through the generation of episodic reactive behaviors, where the Transformer architecture has been proven to be the state-of-the-art. However, they still fail to develop reasoning. Recently, a novel approach for developing cognitive architectures, called Synthetic Cognition, has been proposed and implemented to develop instantaneous reactive behavior. In this study, we aim to explore the use of Synthetic Cognition to develop episodic reactive behaviors. We propose a mechanism to deal with sequences for the recent implementation of Synthetic Cognition, and test it against DNA foundation models in DNA sequence classification tasks. In our experiments, our proposal clearly outperforms the DNA foundation models, obtaining the best score on more benchmark tasks than the alternatives. Thus, we achieve two goals: expanding Synthetic Cognition to deal with sequences, and beating the Transformer architecture for sequence classification.
- [98] arXiv:2504.07664 (cross-list from cs.SE) [pdf, html, other]
-
Title: Data Requirement Goal Modeling for Machine Learning SystemsSubjects: Software Engineering (cs.SE); Machine Learning (cs.LG)
Machine Learning (ML) has been integrated into various software and systems. Two main components are essential for training an ML model: the training data and the ML algorithm. Given the critical role of data in ML system development, it has become increasingly important to assess the quality of data attributes and ensure that the data meets specific requirements before its utilization. This work proposes an approach to guide non-experts in identifying data requirements for ML systems using goal modeling. In this approach, we first develop the Data Requirement Goal Model (DRGM) by surveying the white literature to identify and categorize the issues and challenges faced by data scientists and requirement engineers working on ML-related projects. An initial DRGM was built to accommodate common tasks that would generalize across projects. Then, based on insights from both white and gray literature, a customization mechanism is built to help adjust the tasks, KPIs, and goals' importance of different elements within the DRGM. The generated model can aid its users in evaluating different datasets using GRL evaluation strategies. We then validate the approach through two illustrative examples based on real-world projects. The results from the illustrative examples demonstrate that the data requirements identified by the proposed approach align with the requirements of real-world projects, demonstrating the practicality and effectiveness of the proposed framework. The proposed dataset selection customization mechanism and the proposed DRGM are helpful in guiding non-experts in identifying the data requirements for machine learning systems tailored to a specific ML problem. This approach also aids in evaluating different dataset alternatives to choose the optimum dataset for the problem. For future work, we recommend implementing tool support to generate the DRGM based on a chatbot interface.
- [99] arXiv:2504.07696 (cross-list from eess.IV) [pdf, html, other]
-
Title: Conformalized Generative Bayesian Imaging: An Uncertainty Quantification Framework for Computational ImagingComments: 19 pages, 9 figures, preprintSubjects: Image and Video Processing (eess.IV); Machine Learning (cs.LG)
Uncertainty quantification plays an important role in achieving trustworthy and reliable learning-based computational imaging. Recent advances in generative modeling and Bayesian neural networks have enabled the development of uncertainty-aware image reconstruction methods. Current generative model-based methods seek to quantify the inherent (aleatoric) uncertainty on the underlying image for given measurements by learning to sample from the posterior distribution of the underlying image. On the other hand, Bayesian neural network-based approaches aim to quantify the model (epistemic) uncertainty on the parameters of a deep neural network-based reconstruction method by approximating the posterior distribution of those parameters. Unfortunately, an ongoing need for an inversion method that can jointly quantify complex aleatoric uncertainty and epistemic uncertainty patterns still persists. In this paper, we present a scalable framework that can quantify both aleatoric and epistemic uncertainties. The proposed framework accepts an existing generative model-based posterior sampling method as an input and introduces an epistemic uncertainty quantification capability through Bayesian neural networks with latent variables and deep ensembling. Furthermore, by leveraging the conformal prediction methodology, the proposed framework can be easily calibrated to ensure rigorous uncertainty quantification. We evaluated the proposed framework on magnetic resonance imaging, computed tomography, and image inpainting problems and showed that the epistemic and aleatoric uncertainty estimates produced by the proposed framework display the characteristic features of true epistemic and aleatoric uncertainties. Furthermore, our results demonstrated that the use of conformal prediction on top of the proposed framework enables marginal coverage guarantees consistent with frequentist principles.
- [100] arXiv:2504.07726 (cross-list from cs.DL) [pdf, html, other]
-
Title: Quantum Machine Learning: Unveiling Trends, Impacts through Bibliometric AnalysisSubjects: Digital Libraries (cs.DL); Machine Learning (cs.LG)
Quantum Machine Learning (QML) is the intersection of two revolutionary fields: quantum computing and machine learning. It promises to unlock unparalleled capabilities in data analysis, model building, and problem-solving by harnessing the unique properties of quantum mechanics. This research endeavors to conduct a comprehensive bibliometric analysis of scientific information pertaining to QML covering the period from 2000 to 2023. An extensive dataset comprising 9493 scholarly works is meticulously examined to unveil notable trends, impact factors, and funding patterns within the domain. Additionally, the study employs bibliometric mapping techniques to visually illustrate the network relationships among key countries, institutions, authors, patent citations and significant keywords in QML research. The analysis reveals a consistent growth in publications over the examined period. The findings highlight the United States and China as prominent contributors, exhibiting substantial publication and citation metrics. Notably, the study concludes that QML, as a research subject, is currently in a formative stage, characterized by robust scholarly activity and ongoing development.
- [101] arXiv:2504.07736 (cross-list from physics.flu-dyn) [pdf, html, other]
-
Title: A Novel Deep Learning Approach for Emulating Computationally Expensive Postfire Debris FlowsComments: Manuscript submitted to Computers & Geosciences, 22 pages, 10 figuresSubjects: Fluid Dynamics (physics.flu-dyn); Machine Learning (cs.LG); Geophysics (physics.geo-ph)
Traditional physics-based models of geophysical flows, such as debris flows and landslides that pose significant risks to human lives and infrastructure are computationally expensive, limiting their utility for large-scale parameter sweeps, uncertainty quantification, inversions or real-time applications. This study presents an efficient alternative, a deep learning-based surrogate model built using a modified U-Net architecture to predict the dynamics of runoff-generated debris flows across diverse terrain based on data from physics based simulations. The study area is divided into smaller patches for localized predictions using a patch-predict-stitch methodology (complemented by limited global data to accelerate training). The patches are then combined to reconstruct spatially continuous flow maps, ensuring scalability for large domains. To enable fast training using limited expensive simulations, the deep learning model was trained on data from an ensemble of physics based simulations using parameters generated via Latin Hypercube Sampling and validated on unseen parameter sets and terrain, achieving maximum pointwise errors below 10% and robust generalization. Uncertainty quantification using Monte Carlo methods are enabled using the validated surrogate, which can facilitate probabilistic hazard assessments. This study highlights the potential of deep learning surrogates as powerful tools for geophysical flow analysis, enabling computationally efficient and reliable probabilistic hazard map predictions.
- [102] arXiv:2504.07741 (cross-list from physics.flu-dyn) [pdf, html, other]
-
Title: Harnessing Equivariance: Modeling Turbulence with Graph Neural NetworksComments: 17 pages, 10 figuresSubjects: Fluid Dynamics (physics.flu-dyn); Machine Learning (cs.LG)
This work proposes a novel methodology for turbulence modeling in Large Eddy Simulation (LES) based on Graph Neural Networks (GNNs), which embeds the discrete rotational, reflectional and translational symmetries of the Navier-Stokes equations into the model architecture. In addition, suitable invariant input and output spaces are derived that allow the GNN models to be embedded seamlessly into the LES framework to obtain a symmetry-preserving simulation setup. The suitability of the proposed approach is investigated for two canonical test cases: Homogeneous Isotropic Turbulence (HIT) and turbulent channel flow. For both cases, GNN models are trained successfully in actual simulations using Reinforcement Learning (RL) to ensure that the models are consistent with the underlying LES formulation and discretization. It is demonstrated for the HIT case that the resulting GNN-based LES scheme recovers rotational and reflectional equivariance up to machine precision in actual simulations. At the same time, the stability and accuracy remain on par with non-symmetry-preserving machine learning models that fail to obey these properties. The same modeling strategy translates well to turbulent channel flow, where the GNN model successfully learns the more complex flow physics and is able to recover the turbulent statistics and Reynolds stresses. It is shown that the GNN model learns a zonal modeling strategy with distinct behaviors in the near-wall and outer regions. The proposed approach thus demonstrates the potential of GNNs for turbulence modeling, especially in the context of LES and RL.
- [103] arXiv:2504.07742 (cross-list from stat.ML) [pdf, html, other]
-
Title: Gradient-based Sample Selection for Faster Bayesian OptimizationSubjects: Machine Learning (stat.ML); Machine Learning (cs.LG)
Bayesian optimization (BO) is an effective technique for black-box optimization. However, its applicability is typically limited to moderate-budget problems due to the cubic complexity in computing the Gaussian process (GP) surrogate model. In large-budget scenarios, directly employing the standard GP model faces significant challenges in computational time and resource requirements. In this paper, we propose a novel approach, gradient-based sample selection Bayesian Optimization (GSSBO), to enhance the computational efficiency of BO. The GP model is constructed on a selected set of samples instead of the whole dataset. These samples are selected by leveraging gradient information to maintain diversity and representation. We provide a theoretical analysis of the gradient-based sample selection strategy and obtain explicit sublinear regret bounds for our proposed framework. Extensive experiments on synthetic and real-world tasks demonstrate that our approach significantly reduces the computational cost of GP fitting in BO while maintaining optimization performance comparable to baseline methods.
- [104] arXiv:2504.07757 (cross-list from cs.AI) [pdf, html, other]
-
Title: Search-contempt: a hybrid MCTS algorithm for training AlphaZero-like engines with better computational efficiencySubjects: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
AlphaZero in 2017 was able to master chess and other games without human knowledge by playing millions of games against itself (self-play), with a computation budget running in the tens of millions of dollars. It used a variant of the Monte Carlo Tree Search (MCTS) algorithm, known as PUCT. This paper introduces search-contempt, a novel hybrid variant of the MCTS algorithm that fundamentally alters the distribution of positions generated in self-play, preferring more challenging positions. In addition, search-contempt has been shown to give a big boost in strength for engines in Odds Chess (where one side receives an unfavorable position from the start). More significantly, it opens up the possibility of training a self-play based engine, in a much more computationally efficient manner with the number of training games running into hundreds of thousands, costing tens of thousands of dollars (instead of tens of millions of training games costing millions of dollars required by AlphaZero). This means that it may finally be possible to train such a program from zero on a standard consumer GPU even with a very limited compute, cost, or time budget.
- [105] arXiv:2504.07763 (cross-list from cs.CY) [pdf, other]
-
Title: Data over dialogue: Why artificial intelligence is unlikely to humanise medicineJournal-ref: Monash University, 2024Subjects: Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC); Machine Learning (cs.LG)
Recently, a growing number of experts in artificial intelligence (AI) and medicine have be-gun to suggest that the use of AI systems, particularly machine learning (ML) systems, is likely to humanise the practice of medicine by substantially improving the quality of clinician-patient relationships. In this thesis, however, I argue that medical ML systems are more likely to negatively impact these relationships than to improve them. In particular, I argue that the use of medical ML systems is likely to comprise the quality of trust, care, empathy, understanding, and communication between clinicians and patients.
- [106] arXiv:2504.07777 (cross-list from astro-ph.IM) [pdf, html, other]
-
Title: Adaptive Detection of Fast Moving Celestial Objects Using a Mixture of Experts and Physical-Inspired Neural NetworkComments: Accepted by the AJSubjects: Instrumentation and Methods for Astrophysics (astro-ph.IM); Earth and Planetary Astrophysics (astro-ph.EP); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG); Optics (physics.optics)
Fast moving celestial objects are characterized by velocities across the celestial sphere that significantly differ from the motions of background stars. In observational images, these objects exhibit distinct shapes, contrasting with the typical appearances of stars. Depending on the observational method employed, these celestial entities may be designated as near-Earth objects or asteroids. Historically, fast moving celestial objects have been observed using ground-based telescopes, where the relative stability of stars and Earth facilitated effective image differencing techniques alongside traditional fast moving celestial object detection and classification algorithms. However, the growing prevalence of space-based telescopes, along with their diverse observational modes, produces images with different properties, rendering conventional methods less effective. This paper presents a novel algorithm for detecting fast moving celestial objects within star fields. Our approach enhances state-of-the-art fast moving celestial object detection neural networks by transforming them into physical-inspired neural networks. These neural networks leverage the point spread function of the telescope and the specific observational mode as prior information; they can directly identify moving fast moving celestial objects within star fields without requiring additional training, thereby addressing the limitations of traditional techniques. Additionally, all neural networks are integrated using the mixture of experts technique, forming a comprehensive fast moving celestial object detection algorithm. We have evaluated our algorithm using simulated observational data that mimics various observations carried out by space based telescope scenarios and real observation images. Results demonstrate that our method effectively detects fast moving celestial objects across different observational modes.
- [107] arXiv:2504.07803 (cross-list from cs.CL) [pdf, other]
-
Title: A System for Comprehensive Assessment of RAG FrameworksComments: Technical Report, 7 pages, 2 figures, 1 tableSubjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Retrieval Augmented Generation (RAG) has emerged as a standard paradigm for enhancing the factual accuracy and contextual relevance of Large Language Models (LLMs) by integrating retrieval mechanisms. However, existing evaluation frameworks fail to provide a holistic black-box approach to assessing RAG systems, especially in real-world deployment scenarios. To address this gap, we introduce SCARF (System for Comprehensive Assessment of RAG Frameworks), a modular and flexible evaluation framework designed to benchmark deployed RAG applications systematically. SCARF provides an end-to-end, black-box evaluation methodology, enabling a limited-effort comparison across diverse RAG frameworks. Our framework supports multiple deployment configurations and facilitates automated testing across vector databases and LLM serving strategies, producing a detailed performance report. Moreover, SCARF integrates practical considerations such as response coherence, providing a scalable and adaptable solution for researchers and industry professionals evaluating RAG applications. Using the REST APIs interface, we demonstrate how SCARF can be applied to real-world scenarios, showcasing its flexibility in assessing different RAG frameworks and configurations. SCARF is available at GitHub repository.
- [108] arXiv:2504.07818 (cross-list from stat.ML) [pdf, other]
-
Title: Performance of Rank-One Tensor Approximation on Incomplete DataSubjects: Machine Learning (stat.ML); Machine Learning (cs.LG); Probability (math.PR)
We are interested in the estimation of a rank-one tensor signal when only a portion $\varepsilon$ of its noisy observation is available. We show that the study of this problem can be reduced to that of a random matrix model whose spectral analysis gives access to the reconstruction performance. These results shed light on and specify the loss of performance induced by an artificial reduction of the memory cost of a tensor via the deletion of a random part of its entries.
- [109] arXiv:2504.07820 (cross-list from stat.ML) [pdf, other]
-
Title: Smoothed Distance Kernels for MMDs and Applications in Wasserstein Gradient FlowsComments: 48 pages, 10 figuresSubjects: Machine Learning (stat.ML); Machine Learning (cs.LG); Functional Analysis (math.FA); Probability (math.PR)
Negative distance kernels $K(x,y) := - \|x-y\|$ were used in the definition of maximum mean discrepancies (MMDs) in statistics and lead to favorable numerical results in various applications. In particular, so-called slicing techniques for handling high-dimensional kernel summations profit from the simple parameter-free structure of the distance kernel. However, due to its non-smoothness in $x=y$, most of the classical theoretical results, e.g. on Wasserstein gradient flows of the corresponding MMD functional do not longer hold true. In this paper, we propose a new kernel which keeps the favorable properties of the negative distance kernel as being conditionally positive definite of order one with a nearly linear increase towards infinity and a simple slicing structure, but is Lipschitz differentiable now. Our construction is based on a simple 1D smoothing procedure of the absolute value function followed by a Riemann-Liouville fractional integral transform. Numerical results demonstrate that the new kernel performs similarly well as the negative distance kernel in gradient descent methods, but now with theoretical guarantees.
- [110] arXiv:2504.07898 (cross-list from cs.IR) [pdf, html, other]
-
Title: How do Large Language Models Understand Relevance? A Mechanistic Interpretability PerspectiveSubjects: Information Retrieval (cs.IR); Computation and Language (cs.CL); Machine Learning (cs.LG)
Recent studies have shown that large language models (LLMs) can assess relevance and support information retrieval (IR) tasks such as document ranking and relevance judgment generation. However, the internal mechanisms by which off-the-shelf LLMs understand and operationalize relevance remain largely unexplored. In this paper, we systematically investigate how different LLM modules contribute to relevance judgment through the lens of mechanistic interpretability. Using activation patching techniques, we analyze the roles of various model components and identify a multi-stage, progressive process in generating either pointwise or pairwise relevance judgment. Specifically, LLMs first extract query and document information in the early layers, then process relevance information according to instructions in the middle layers, and finally utilize specific attention heads in the later layers to generate relevance judgments in the required format. Our findings provide insights into the mechanisms underlying relevance assessment in LLMs, offering valuable implications for future research on leveraging LLMs for IR tasks.
- [111] arXiv:2504.07904 (cross-list from eess.IV) [pdf, html, other]
-
Title: The Efficacy of Semantics-Preserving Transformations in Self-Supervised Learning for Medical UltrasoundComments: 17 pages, 12 figures, 18 tables, Submitted to Medical Image AnalysisSubjects: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
Data augmentation is a central component of joint embedding self-supervised learning (SSL). Approaches that work for natural images may not always be effective in medical imaging tasks. This study systematically investigated the impact of data augmentation and preprocessing strategies in SSL for lung ultrasound. Three data augmentation pipelines were assessed: (1) a baseline pipeline commonly used across imaging domains, (2) a novel semantic-preserving pipeline designed for ultrasound, and (3) a distilled set of the most effective transformations from both pipelines. Pretrained models were evaluated on multiple classification tasks: B-line detection, pleural effusion detection, and COVID-19 classification. Experiments revealed that semantics-preserving data augmentation resulted in the greatest performance for COVID-19 classification - a diagnostic task requiring global image context. Cropping-based methods yielded the greatest performance on the B-line and pleural effusion object classification tasks, which require strong local pattern recognition. Lastly, semantics-preserving ultrasound image preprocessing resulted in increased downstream performance for multiple tasks. Guidance regarding data augmentation and preprocessing strategies was synthesized for practitioners working with SSL in ultrasound.
- [112] arXiv:2504.07923 (cross-list from q-fin.TR) [pdf, html, other]
-
Title: Trading Graph Neural NetworkSubjects: Trading and Market Microstructure (q-fin.TR); Machine Learning (cs.LG); General Economics (econ.GN); Pricing of Securities (q-fin.PR)
This paper proposes a new algorithm -- Trading Graph Neural Network (TGNN) that can structurally estimate the impact of asset features, dealer features and relationship features on asset prices in trading networks. It combines the strength of the traditional simulated method of moments (SMM) and recent machine learning techniques -- Graph Neural Network (GNN). It outperforms existing reduced-form methods with network centrality measures in prediction accuracy. The method can be used on networks with any structure, allowing for heterogeneity among both traders and assets.
Cross submissions (showing 55 of 55 entries)
- [113] arXiv:2303.08431 (replaced) [pdf, html, other]
-
Title: Policy Gradient Converges to the Globally Optimal Policy for Nearly Linear-Quadratic RegulatorsComments: 34 pagesSubjects: Machine Learning (cs.LG); Optimization and Control (math.OC); Machine Learning (stat.ML)
Nonlinear control systems with partial information to the decision maker are prevalent in a variety of applications. As a step toward studying such nonlinear systems, this work explores reinforcement learning methods for finding the optimal policy in the nearly linear-quadratic regulator systems. In particular, we consider a dynamic system that combines linear and nonlinear components, and is governed by a policy with the same structure. Assuming that the nonlinear component comprises kernels with small Lipschitz coefficients, we characterize the optimization landscape of the cost function. Although the cost function is nonconvex in general, we establish the local strong convexity and smoothness in the vicinity of the global optimizer. Additionally, we propose an initialization mechanism to leverage these properties. Building on the developments, we design a policy gradient algorithm that is guaranteed to converge to the globally optimal policy with a linear rate.
- [114] arXiv:2305.10449 (replaced) [pdf, html, other]
-
Title: Cooperation Is All You NeedAhsan Adeel, Junaid Muzaffar, Khubaib Ahmed, Mohsin Raza, Fahad Zia, Eamin Chaudary, Talha Bin Riaz, Ahmed SaeedSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Neural and Evolutionary Computing (cs.NE)
Going beyond 'dendritic democracy', we introduce a 'democracy of local processors', termed Cooperator. Here we compare their capabilities when used in permutation invariant neural networks for reinforcement learning (RL), with machine learning algorithms based on Transformers, such as ChatGPT. Transformers are based on the long standing conception of integrate-and-fire 'point' neurons, whereas Cooperator is inspired by recent neurobiological breakthroughs suggesting that the cellular foundations of mental life depend on context-sensitive pyramidal neurons in the neocortex which have two functionally distinct points. Weshow that when used for RL, an algorithm based on Cooperator learns far quicker than that based on Transformer, even while having the same number of parameters.
- [115] arXiv:2307.06162 (replaced) [pdf, other]
-
Title: Deep Generative Models for Physiological Signals: A Systematic Literature ReviewComments: accepted in Elsevier Artificial Intelligence in Medicine, 38 pagesJournal-ref: Elsevier Artificial Intelligence in Medicine, 2025Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Signal Processing (eess.SP)
In this paper, we present a systematic literature review on deep generative models for physiological signals, particularly electrocardiogram (ECG), electroencephalogram (EEG), photoplethysmogram (PPG) and electromyogram (EMG). Compared to the existing review papers, we present the first review that summarizes the recent state-of-the-art deep generative models. By analyzing the state-of-the-art research related to deep generative models along with their main applications and challenges, this review contributes to the overall understanding of these models applied to physiological signals. Additionally, by highlighting the employed evaluation protocol and the most used physiological databases, this review facilitates the assessment and benchmarking of deep generative models.
- [116] arXiv:2309.00508 (replaced) [pdf, html, other]
-
Title: Geometry and Local Recovery of Global Minima of Two-layer Neural Networks at OverparameterizationComments: Some typos about separating inputs are fixedSubjects: Machine Learning (cs.LG); Dynamical Systems (math.DS)
Under mild assumptions, we investigate the geometry of the loss landscape for two-layer neural networks in the vicinity of global minima. Utilizing novel techniques, we demonstrate: (i) how global minima with zero generalization error become geometrically separated from other global minima as the sample size grows; and (ii) the local convergence properties and rate of gradient flow dynamics. Our results indicate that two-layer neural networks can be locally recovered in the regime of overparameterization.
- [117] arXiv:2401.13360 (replaced) [pdf, html, other]
-
Title: Understanding and Mitigating the Bias in Sample Selection for Learning with Noisy LabelsSubjects: Machine Learning (cs.LG)
Learning with noisy labels aims to ensure model generalization given a label-corrupted training set. The sample selection strategy achieves promising performance by selecting a label-reliable subset for model training. In this paper, we empirically reveal that existing sample selection methods suffer from both data and training bias that are represented as imbalanced selected sets and accumulation errors in practice, respectively. However, only the training bias was handled in previous studies. To address this limitation, we propose a noIse-Tolerant Expert Model (ITEM) for debiased learning in sample selection. Specifically, to mitigate the training bias, we design a robust network architecture that integrates with multiple experts. Compared with the prevailing double-branch network, our network exhibits better performance of selection and prediction by ensembling these experts while training with fewer parameters. Meanwhile, to mitigate the data bias, we propose a mixed sampling strategy based on two weight-based data samplers. By training on the mixture of two class-discriminative mini-batches, the model mitigates the effect of the imbalanced training set while avoiding sparse representations that are easily caused by sampling strategies. Extensive experiments and analyses demonstrate the effectiveness of ITEM. Our code is available at this url \href{this https URL}{ITEM}.
- [118] arXiv:2402.06038 (replaced) [pdf, html, other]
-
Title: Understanding Contrastive Representation Learning from Positive Unlabeled (PU) DataAnish Acharya, Li Jing, Bhargav Bhushanam, Dhruv Choudhary, Michael Rabbat, Sujay Sanghavi, Inderjit S DhillonSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
Pretext Invariant Representation Learning (PIRL) followed by Supervised Fine-Tuning (SFT) has become a standard paradigm for learning with limited labels. We extend this approach to the Positive Unlabeled (PU) setting, where only a small set of labeled positives and a large unlabeled pool -- containing both positives and negatives are available. We study this problem under two regimes: (i) without access to the class prior, and (ii) when the prior is known or can be estimated. We introduce Positive Unlabeled Contrastive Learning (puCL), an unbiased and variance reducing contrastive objective that integrates weak supervision from labeled positives judiciously into the contrastive loss. When the class prior is known, we propose Positive Unlabeled InfoNCE (puNCE), a prior-aware extension that re-weights unlabeled samples as soft positive negative mixtures. For downstream classification, we develop a pseudo-labeling algorithm that leverages the structure of the learned embedding space via PU aware clustering. Our framework is supported by theory; offering bias-variance analysis, convergence insights, and generalization guarantees via augmentation concentration; and validated empirically across standard PU benchmarks, where it consistently outperforms existing methods, particularly in low-supervision regimes.
- [119] arXiv:2405.01814 (replaced) [pdf, html, other]
-
Title: Efficient Heterogeneous Large Language Model Decoding with Model-Attention DisaggregationShaoyuan Chen, Wencong Xiao, Yutong Lin, Mingxing Zhang, Yingdi Shan, Jinlei Jiang, Kang Chen, Yongwei WuSubjects: Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC)
Transformer-based large language models (LLMs) exhibit impressive performance in generative tasks but also introduce significant challenges in real-world serving due to inefficient use of the expensive, computation-optimized accelerators. Although disaggregated serving architectures have been proposed to split different phases of LLM inference, the efficiency of decoding phase is still low. This is caused by the varying resource demands of different operators in the transformer-based LLMs. Specifically, the attention operator is memory-intensive, exhibiting a memory access pattern that clashes with the strengths of modern accelerators, especially for long context requests. To enhance the efficiency of LLM decoding, we introduce model-attention disaggregation. This approach leverages a collection of cheap, memory-optimized devices for the attention operator while still utilizing high-end accelerators for other parts of the model. This heterogeneous setup ensures that each component is tailored to its specific workload, maximizing overall performance and cost efficiency. Our comprehensive analysis and experiments confirm the viability of splitting the attention computation over multiple devices. Also, the communication bandwidth required between heterogeneous devices proves to be manageable with prevalent networking technologies. To further validate our theory, we develop and deploy Lamina, an LLM inference system that incorporates model-attention disaggregation in a distributed heterogeneous cluster. Experimental results indicate that Lamina can provide 16.1 ~ 90.1% higher estimated throughput than existing solutions with similar costs.
- [120] arXiv:2405.04710 (replaced) [pdf, html, other]
-
Title: Untangling Lariats: Subgradient Following of Variationally Penalized ObjectivesSubjects: Machine Learning (cs.LG); Optimization and Control (math.OC)
We describe an apparatus for subgradient-following of the optimum of convex problems with variational penalties. In this setting, we receive a sequence $y_i,\ldots,y_n$ and seek a smooth sequence $x_1,\ldots,x_n$. The smooth sequence needs to attain the minimum Bregman divergence to an input sequence with additive variational penalties in the general form of $\sum_i{}g_i(x_{i+1}-x_i)$. We derive known algorithms such as the fused lasso and isotonic regression as special cases of our approach. Our approach also facilitates new variational penalties such as non-smooth barrier functions.
We then derive a novel lattice-based procedure for subgradient following of variational penalties characterized through the output of arbitrary convolutional filters. This paradigm yields efficient solvers for high-order filtering problems of temporal sequences in which sparse discrete derivatives such as acceleration and jerk are desirable. We also introduce and analyze new multivariate problems in which $\mathbf{x}_i,\mathbf{y}_i\in\mathbb{R}^d$ with variational penalties that depend on $\|\mathbf{x}_{i+1}-\mathbf{x}_i\|$. The norms we consider are $\ell_2$ and $\ell_\infty$ which promote group sparsity. - [121] arXiv:2405.08036 (replaced) [pdf, other]
-
Title: POWQMIX: Weighted Value Factorization with Potentially Optimal Joint Actions Recognition for Cooperative Multi-Agent Reinforcement LearningComments: This paper needs further refinementSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Value function factorization methods are commonly used in cooperative multi-agent reinforcement learning, with QMIX receiving significant attention. Many QMIX-based methods introduce monotonicity constraints between the joint action value and individual action values to achieve decentralized execution. However, such constraints limit the representation capacity of value factorization, restricting the joint action values it can represent and hindering the learning of the optimal policy. To address this challenge, we propose the Potentially Optimal Joint Actions Weighted QMIX (POWQMIX) algorithm, which recognizes the potentially optimal joint actions and assigns higher weights to the corresponding losses of these joint actions during training. We theoretically prove that with such a weighted training approach the optimal policy is guaranteed to be recovered. Experiments in matrix games, difficulty-enhanced predator-prey, and StarCraft II Multi-Agent Challenge environments demonstrate that our algorithm outperforms the state-of-the-art value-based multi-agent reinforcement learning methods.
- [122] arXiv:2405.16924 (replaced) [pdf, html, other]
-
Title: Demystifying amortized causal discovery with transformersSubjects: Machine Learning (cs.LG); Machine Learning (stat.ML)
Supervised learning approaches for causal discovery from observational data often achieve competitive performance despite seemingly avoiding explicit assumptions that traditional methods make for identifiability. In this work, we investigate CSIvA (Ke et al., 2023), a transformer-based model promising to train on synthetic data and transfer to real data. First, we bridge the gap with existing identifiability theory and show that constraints on the training data distribution implicitly define a prior on the test observations. Consistent with classical approaches, good performance is achieved when we have a good prior on the test data, and the underlying model is identifiable. At the same time, we find new trade-offs. Training on datasets generated from different classes of causal models, unambiguously identifiable in isolation, improves the test generalization. Performance is still guaranteed, as the ambiguous cases resulting from the mixture of identifiable causal models are unlikely to occur (which we formally prove). Overall, our study finds that amortized causal discovery still needs to obey identifiability theory, but it also differs from classical methods in how the assumptions are formulated, trading more reliance on assumptions on the noise type for fewer hypotheses on the mechanisms.
- [123] arXiv:2406.12816 (replaced) [pdf, html, other]
-
Title: Neural Approximate Mirror Maps for Constrained Diffusion ModelsComments: ICLR 2025Subjects: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
Diffusion models excel at creating visually-convincing images, but they often struggle to meet subtle constraints inherent in the training data. Such constraints could be physics-based (e.g., satisfying a PDE), geometric (e.g., respecting symmetry), or semantic (e.g., including a particular number of objects). When the training data all satisfy a certain constraint, enforcing this constraint on a diffusion model makes it more reliable for generating valid synthetic data and solving constrained inverse problems. However, existing methods for constrained diffusion models are restricted in the constraints they can handle. For instance, recent work proposed to learn mirror diffusion models (MDMs), but analytical mirror maps only exist for convex constraints and can be challenging to derive. We propose neural approximate mirror maps (NAMMs) for general, possibly non-convex constraints. Our approach only requires a differentiable distance function from the constraint set. We learn an approximate mirror map that transforms data into an unconstrained space and a corresponding approximate inverse that maps data back to the constraint set. A generative model, such as an MDM, can then be trained in the learned mirror space and its samples restored to the constraint set by the inverse map. We validate our approach on a variety of constraints, showing that compared to an unconstrained diffusion model, a NAMM-based MDM substantially improves constraint satisfaction. We also demonstrate how existing diffusion-based inverse-problem solvers can be easily applied in the learned mirror space to solve constrained inverse problems.
- [124] arXiv:2406.14528 (replaced) [pdf, html, other]
-
Title: DeciMamba: Exploring the Length Extrapolation Potential of MambaAssaf Ben-Kish, Itamar Zimerman, Shady Abu-Hussein, Nadav Cohen, Amir Globerson, Lior Wolf, Raja GiryesComments: Official Implementation: this https URLSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Long-range sequence processing poses a significant challenge for Transformers due to their quadratic complexity in input length. A promising alternative is Mamba, which demonstrates high performance and achieves Transformer-level capabilities while requiring substantially fewer computational resources. In this paper we explore the length-generalization capabilities of Mamba, which we find to be relatively limited. Through a series of visualizations and analyses we identify that the limitations arise from a restricted effective receptive field, dictated by the sequence length used during training. To address this constraint, we introduce DeciMamba, a context-extension method specifically designed for Mamba. This mechanism, built on top of a hidden filtering mechanism embedded within the S6 layer, enables the trained model to extrapolate well even without additional training. Empirical experiments over real-world long-range NLP tasks show that DeciMamba can extrapolate to context lengths that are significantly longer than the ones seen during training, while enjoying faster inference.
- [125] arXiv:2406.15156 (replaced) [pdf, html, other]
-
Title: Reconsidering Faithfulness in Regular, Self-Explainable and Domain Invariant GNNsComments: Uploading ICLR25 camera ready versionSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
As Graph Neural Networks (GNNs) become more pervasive, it becomes paramount to build reliable tools for explaining their predictions. A core desideratum is that explanations are \textit{faithful}, \ie that they portray an accurate picture of the GNN's reasoning process. However, a number of different faithfulness metrics exist, begging the question of what is faithfulness exactly and how to achieve it. We make three key contributions. We begin by showing that \textit{existing metrics are not interchangeable} -- \ie explanations attaining high faithfulness according to one metric may be unfaithful according to others -- and can systematically ignore important properties of explanations. We proceed to show that, surprisingly, \textit{optimizing for faithfulness is not always a sensible design goal}. Specifically, we prove that for injective regular GNN architectures, perfectly faithful explanations are completely uninformative. This does not apply to modular GNNs, such as self-explainable and domain-invariant architectures, prompting us to study the relationship between architectural choices and faithfulness. Finally, we show that \textit{faithfulness is tightly linked to out-of-distribution generalization}, in that simply ensuring that a GNN can correctly recognize the domain-invariant subgraph, as prescribed by the literature, does not guarantee that it is invariant unless this subgraph is also this http URL code is publicly available on GitHub
- [126] arXiv:2407.04752 (replaced) [pdf, html, other]
-
Title: SpikeLLM: Scaling up Spiking Neural Network to Large Language Models via Saliency-based SpikingSubjects: Machine Learning (cs.LG); Computation and Language (cs.CL); Neural and Evolutionary Computing (cs.NE)
Recent advancements in large language models (LLMs) with billions of parameters have improved performance in various applications, but their inference processes demand significant energy and computational resources. In contrast, the human brain, with approximately 86 billion neurons, is much more energy-efficient than LLMs with similar parameters. Inspired by this, we redesign 7$\sim$70 billion parameter LLMs using bio-plausible spiking mechanisms, emulating the efficient behavior of the human brain. We propose the first spiking large language model, SpikeLLM. Coupled with the proposed model, two essential approaches are proposed to improve spike training efficiency: Generalized Integrate-and-Fire (GIF) neurons to compress spike length from $T$ to $\frac{T}{L} \log_2 L$ bits, and an Optimal Brain Spiking framework to divide outlier channels and allocate different $T$ for GIF neurons, which further compresses spike length to approximate $log_2T$ bits. The necessity of spike-driven LLM is proved by comparison with quantized LLMs with similar operations. In the OmniQuant pipeline, SpikeLLM reduces 11.01% WikiText2 perplexity and improves 2.55% accuracy of common scene reasoning on a LLAMA-7B W4A4 model. In the GPTQ pipeline, SpikeLLM achieves direct additive in linear layers, significantly exceeding PB-LLMs.
- [127] arXiv:2407.08169 (replaced) [pdf, other]
-
Title: The Approximate Fisher Influence Function: Faster Estimation of Data Influence in Statistical ModelsSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Quantifying the influence of infinitesimal changes in training data on model performance is crucial for understanding and improving machine learning models. In this work, we reformulate this problem as a weighted empirical risk minimization and enhance existing influence function-based methods by using information geometry to derive a new algorithm to estimate influence. Our formulation proves versatile across various applications, and we further demonstrate in simulations how it remains informative even in non-convex cases. Furthermore, we show that our method offers significant computational advantages over current Newton step-based methods.
- [128] arXiv:2409.18025 (replaced) [pdf, html, other]
-
Title: An Adversarial Perspective on Machine Unlearning for AI SafetyComments: Final version published in Transactions on Machine Learning Research (TMLR); Best technical paper at Neurips 2024 SoLaR workshopSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Cryptography and Security (cs.CR)
Large language models are finetuned to refuse questions about hazardous knowledge, but these protections can often be bypassed. Unlearning methods aim at completely removing hazardous capabilities from models and make them inaccessible to adversaries. This work challenges the fundamental differences between unlearning and traditional safety post-training from an adversarial perspective. We demonstrate that existing jailbreak methods, previously reported as ineffective against unlearning, can be successful when applied carefully. Furthermore, we develop a variety of adaptive methods that recover most supposedly unlearned capabilities. For instance, we show that finetuning on 10 unrelated examples or removing specific directions in the activation space can recover most hazardous capabilities for models edited with RMU, a state-of-the-art unlearning method. Our findings challenge the robustness of current unlearning approaches and question their advantages over safety training.
- [129] arXiv:2410.02113 (replaced) [pdf, html, other]
-
Title: Mamba Neural Operator: Who Wins? Transformers vs. State-Space Models for PDEsChun-Wun Cheng, Jiahao Huang, Yi Zhang, Guang Yang, Carola-Bibiane Schönlieb, Angelica I Aviles-RiveroSubjects: Machine Learning (cs.LG); Numerical Analysis (math.NA)
Partial differential equations (PDEs) are widely used to model complex physical systems, but solving them efficiently remains a significant challenge. Recently, Transformers have emerged as the preferred architecture for PDEs due to their ability to capture intricate dependencies. However, they struggle with representing continuous dynamics and long-range interactions. To overcome these limitations, we introduce the Mamba Neural Operator (MNO), a novel framework that enhances neural operator-based techniques for solving PDEs. MNO establishes a formal theoretical connection between structured state-space models (SSMs) and neural operators, offering a unified structure that can adapt to diverse architectures, including Transformer-based models. By leveraging the structured design of SSMs, MNO captures long-range dependencies and continuous dynamics more effectively than traditional Transformers. Through extensive analysis, we show that MNO significantly boosts the expressive power and accuracy of neural operators, making it not just a complement but a superior framework for PDE-related tasks, bridging the gap between efficient representation and accurate solution approximation.
- [130] arXiv:2410.04639 (replaced) [pdf, other]
-
Title: Radial Basis Operator NetworksSubjects: Machine Learning (cs.LG)
Operator networks are designed to approximate nonlinear operators, which provide mappings between infinite-dimensional spaces such as function spaces. These networks are playing an increasingly important role in machine learning, with their most notable contributions in the field of scientific computing. Their significance stems from their ability to handle the type of data often encountered in scientific applications. For instance, in climate modeling or fluid dynamics, input data typically consists of discretized continuous fields (like temperature distributions or velocity fields). We introduce the radial basis operator network (RBON), which represents a significant advancement as the first operator network capable of learning an operator in both the time domain and frequency domain when adjusted to accept complex-valued inputs. Despite the small, single hidden-layer structure, the RBON boasts small $L^2$ relative test error for both in- and out-of-distribution data (OOD) of less than $1\times 10^{-7}$ in some benchmark cases. Moreover, the RBON maintains small error on OOD data from entirely different function classes from the training data.
- [131] arXiv:2410.06264 (replaced) [pdf, other]
-
Title: Think While You Generate: Discrete Diffusion with Planned DenoisingSulin Liu, Juno Nam, Andrew Campbell, Hannes Stärk, Yilun Xu, Tommi Jaakkola, Rafael Gómez-BombarelliComments: ICLR 2025Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (stat.ML)
Discrete diffusion has achieved state-of-the-art performance, outperforming or approaching autoregressive models on standard benchmarks. In this work, we introduce Discrete Diffusion with Planned Denoising (DDPD), a novel framework that separates the generation process into two models: a planner and a denoiser. At inference time, the planner selects which positions to denoise next by identifying the most corrupted positions in need of denoising, including both initially corrupted and those requiring additional refinement. This plan-and-denoise approach enables more efficient reconstruction during generation by iteratively identifying and denoising corruptions in the optimal order. DDPD outperforms traditional denoiser-only mask diffusion methods, achieving superior results on language modeling benchmarks such as text8, OpenWebText, and token-based image generation on ImageNet $256 \times 256$. Notably, in language modeling, DDPD significantly reduces the performance gap between diffusion-based and autoregressive methods in terms of generative perplexity. Code is available at this https URL.
- [132] arXiv:2410.08709 (replaced) [pdf, html, other]
-
Title: Distillation of Discrete Diffusion through Dimensional CorrelationsComments: 39 pages, GitHub link addedSubjects: Machine Learning (cs.LG); Numerical Analysis (math.NA); Machine Learning (stat.ML)
Diffusion models have demonstrated exceptional performances in various fields of generative modeling, but suffer from slow sampling speed due to their iterative nature. While this issue is being addressed in continuous domains, discrete diffusion models face unique challenges, particularly in capturing dependencies between elements (e.g., pixel relationships in image, sequential dependencies in language) mainly due to the computational cost of processing high-dimensional joint distributions. In this paper, (i) we propose "mixture" models for discrete diffusion that are capable of treating dimensional correlations while remaining scalable, and (ii) we provide a set of loss functions for distilling the iterations of existing models. Two primary theoretical insights underpin our approach: First, conventional models with element-wise independence can well approximate the data distribution, but essentially require {\it many sampling steps}. Second, our loss functions enable the mixture models to distill such many-step conventional models into just a few steps by learning the dimensional correlations. Our experimental results show the effectiveness of the proposed method in distilling pretrained discrete diffusion models across image and language domains. The code used in the paper is available at this https URL .
- [133] arXiv:2410.08893 (replaced) [pdf, html, other]
-
Title: Drama: Mamba-Enabled Model-Based Reinforcement Learning Is Sample and Parameter EfficientComments: Published as a conference paper at ICLR 2025Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Robotics (cs.RO)
Model-based reinforcement learning (RL) offers a solution to the data inefficiency that plagues most model-free RL algorithms. However, learning a robust world model often requires complex and deep architectures, which are computationally expensive and challenging to train. Within the world model, sequence models play a critical role in accurate predictions, and various architectures have been explored, each with its own challenges. Currently, recurrent neural network (RNN)-based world models struggle with vanishing gradients and capturing long-term dependencies. Transformers, on the other hand, suffer from the quadratic memory and computational complexity of self-attention mechanisms, scaling as $O(n^2)$, where $n$ is the sequence length.
To address these challenges, we propose a state space model (SSM)-based world model, Drama, specifically leveraging Mamba, that achieves $O(n)$ memory and computational complexity while effectively capturing long-term dependencies and enabling efficient training with longer sequences. We also introduce a novel sampling method to mitigate the suboptimality caused by an incorrect world model in the early training stages. Combining these techniques, Drama achieves a normalised score on the Atari100k benchmark that is competitive with other state-of-the-art (SOTA) model-based RL algorithms, using only a 7 million-parameter world model. Drama is accessible and trainable on off-the-shelf hardware, such as a standard laptop. Our code is available at this https URL. - [134] arXiv:2410.22318 (replaced) [pdf, html, other]
-
Title: Online Detecting LLM-Generated Texts via Sequential Hypothesis Testing by BettingSubjects: Machine Learning (cs.LG)
Developing algorithms to differentiate between machine-generated texts and human-written texts has garnered substantial attention in recent years. Existing methods in this direction typically concern an offline setting where a dataset containing a mix of real and machine-generated texts is given upfront, and the task is to determine whether each sample in the dataset is from a large language model (LLM) or a human. However, in many practical scenarios, sources such as news websites, social media accounts, or on other forums publish content in a streaming fashion. Therefore, in this online scenario, how to quickly and accurately determine whether the source is an LLM with strong statistical guarantees is crucial for these media or platforms to function effectively and prevent the spread of misinformation and other potential misuse of LLMs. To tackle the problem of online detection, we develop an algorithm based on the techniques of sequential hypothesis testing by betting that not only builds upon and complements existing offline detection techniques but also enjoys statistical guarantees, which include a controlled false positive rate and the expected time to correctly identify a source as an LLM. Experiments were conducted to demonstrate the effectiveness of our method.
- [135] arXiv:2411.05874 (replaced) [pdf, html, other]
-
Title: Interplay between Federated Learning and Explainable Artificial Intelligence: a Scoping ReviewLuis M. Lopez-Ramos, Florian Leiser, Aditya Rastogi, Steven Hicks, Inga Strümke, Vince I. Madai, Tobias Budig, Ali Sunyaev, Adam HilbertComments: 16 pages, 10 figures, submitted in IEEE AccessSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
The joint implementation of federated learning (FL) and explainable artificial intelligence (XAI) could allow training models from distributed data and explaining their inner workings while preserving essential aspects of privacy. Toward establishing the benefits and tensions associated with their interplay, this scoping review maps the publications that jointly deal with FL and XAI, focusing on publications that reported an interplay between FL and model interpretability or post-hoc explanations. Out of the 37 studies meeting our criteria, only one explicitly and quantitatively analyzed the influence of FL on model explanations, revealing a significant research gap. The aggregation of interpretability metrics across FL nodes created generalized global insights at the expense of node-specific patterns being diluted. Several studies proposed FL algorithms incorporating explanation methods to safeguard the learning process against defaulting or malicious nodes. Studies using established FL libraries or following reporting guidelines are a minority. More quantitative research and structured, transparent practices are needed to fully understand their mutual impact and under which conditions it happens.
- [136] arXiv:2412.14602 (replaced) [pdf, html, other]
-
Title: Towards Scalable and Deep Graph Neural Networks via Noise MaskingSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
In recent years, Graph Neural Networks (GNNs) have achieved remarkable success in many graph mining tasks. However, scaling them to large graphs is challenging due to the high computational and storage costs of repeated feature propagation and non-linear transformation during training. One commonly employed approach to address this challenge is model-simplification, which only executes the Propagation (P) once in the pre-processing, and Combine (C) these receptive fields in different ways and then feed them into a simple model for better performance. Despite their high predictive performance and scalability, these methods still face two limitations. First, existing approaches mainly focus on exploring different C methods from the model perspective, neglecting the crucial problem of performance degradation with increasing P depth from the data-centric perspective, known as the over-smoothing problem. Second, pre-processing overhead takes up most of the end-to-end processing time, especially for large-scale graphs. To address these limitations, we present random walk with noise masking (RMask), a plug-and-play module compatible with the existing model-simplification works. This module enables the exploration of deeper GNNs while preserving their scalability. Unlike the previous model-simplification works, we focus on continuous P and found that the noise existing inside each P is the cause of the over-smoothing issue, and use the efficient masking mechanism to eliminate them. Experimental results on six real-world datasets demonstrate that model-simplification works equipped with RMask yield superior performance compared to their original version and can make a good trade-off between accuracy and efficiency.
- [137] arXiv:2412.14865 (replaced) [pdf, other]
-
Title: Hierarchical Subspaces of Policies for Continual Offline Reinforcement LearningSubjects: Machine Learning (cs.LG)
We consider a Continual Reinforcement Learning setup, where a learning agent must continuously adapt to new tasks while retaining previously acquired skill sets, with a focus on the challenge of avoiding forgetting past gathered knowledge and ensuring scalability with the growing number of tasks. Such issues prevail in autonomous robotics and video game simulations, notably for navigation tasks prone to topological or kinematic changes. To address these issues, we introduce HiSPO, a novel hierarchical framework designed specifically for continual learning in navigation settings from offline data. Our method leverages distinct policy subspaces of neural networks to enable flexible and efficient adaptation to new tasks while preserving existing knowledge. We demonstrate, through a careful experimental study, the effectiveness of our method in both classical MuJoCo maze environments and complex video game-like navigation simulations, showcasing competitive performances and satisfying adaptability with respect to classical continual learning metrics, in particular regarding the memory usage and efficiency.
- [138] arXiv:2501.05279 (replaced) [pdf, html, other]
-
Title: Learning convolution operators on compact Abelian groupsSubjects: Machine Learning (cs.LG); Machine Learning (stat.ML)
We consider the problem of learning convolution operators associated to compact Abelian groups. We study a regularization-based approach and provide corresponding learning guarantees under natural regularity conditions on the convolution kernel. More precisely, we assume the convolution kernel is a function in a translation invariant Hilbert space and analyze a natural ridge regression (RR) estimator. Building on existing results for RR, we characterize the accuracy of the estimator in terms of finite sample bounds. Interestingly, regularity assumptions which are classical in the analysis of RR, have a novel and natural interpretation in terms of space/frequency localization. Theoretical results are illustrated by numerical simulations.
- [139] arXiv:2501.05550 (replaced) [pdf, html, other]
-
Title: Emergent weight morphologies in deep neural networksSubjects: Machine Learning (cs.LG); Disordered Systems and Neural Networks (cond-mat.dis-nn)
Whether deep neural networks can exhibit emergent behaviour is not only relevant for understanding how deep learning works, it is also pivotal for estimating potential security risks of increasingly capable artificial intelligence systems. Here, we show that training deep neural networks gives rise to emergent weight morphologies independent of the training data. Specifically, in analogy to condensed matter physics, we derive a theory that predict that the homogeneous state of deep neural networks is unstable in a way that leads to the emergence of periodic channel structures. We verified these structures by performing numerical experiments on a variety of data sets. Our work demonstrates emergence in the training of deep neural networks, which impacts the achievable performance of deep neural networks.
- [140] arXiv:2501.13011 (replaced) [pdf, html, other]
-
Title: MONA: Myopic Optimization with Non-myopic Approval Can Mitigate Multi-step Reward HackingSebastian Farquhar, Vikrant Varma, David Lindner, David Elson, Caleb Biddulph, Ian Goodfellow, Rohin ShahSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Future advanced AI systems may learn sophisticated strategies through reinforcement learning (RL) that humans cannot understand well enough to safely evaluate. We propose a training method which avoids agents learning undesired multi-step plans that receive high reward (multi-step "reward hacks") even if humans are not able to detect that the behaviour is undesired. The method, Myopic Optimization with Non-myopic Approval (MONA), works by combining short-sighted optimization with far-sighted reward. We demonstrate that MONA can prevent multi-step reward hacking that ordinary RL causes, even without being able to detect the reward hacking and without any extra information that ordinary RL does not get access to. We study MONA empirically in three settings which model different misalignment failure modes including 2-step environments with LLMs representing delegated oversight and encoded reasoning and longer-horizon gridworld environments representing sensor tampering.
- [141] arXiv:2501.13230 (replaced) [pdf, html, other]
-
Title: Let SSMs be ConvNets: State-space Modeling with Optimal Tensor ContractionsComments: 25 pages, 7 figuresSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
We introduce Centaurus, a class of networks composed of generalized state-space model (SSM) blocks, where the SSM operations can be treated as tensor contractions during training. The optimal order of tensor contractions can then be systematically determined for every SSM block to maximize training efficiency. This allows more flexibility in designing SSM blocks beyond the depthwise-separable configuration commonly implemented. The new design choices will take inspiration from classical convolutional blocks including group convolutions, full convolutions, and bottleneck blocks. We architect the Centaurus network with a mixture of these blocks, to balance between network size and performance, as well as memory and computational efficiency during both training and inference. We show that this heterogeneous network design outperforms its homogeneous counterparts in raw audio processing tasks including keyword spotting, speech denoising, and automatic speech recognition (ASR). For ASR, Centaurus is the first network with competitive performance that can be made fully state-space based, without using any nonlinear recurrence (LSTMs), explicit convolutions (CNNs), or (surrogate) attention mechanism. The source code is available as supplementary material on this https URL
- [142] arXiv:2501.15293 (replaced) [pdf, other]
-
Title: Deep Learning in Early Alzheimer's disease's Detection: A Comprehensive Survey of Classification, Segmentation, and Feature Extraction MethodsRubab Hafeez, Sadia Waheed, Syeda Aleena Naqvi, Fahad Maqbool, Amna Sarwar, Sajjad Saleem, Muhammad Imran Sharif, Kamran Siddique, Zahid AkhtarComments: 22 pagesSubjects: Machine Learning (cs.LG)
Alzheimers disease is a deadly neurological condition, impairing important memory and brain functions. Alzheimers disease promotes brain shrinkage, ultimately leading to dementia. Dementia diagnosis typically takes 2.8 to 4.4 years after the first clinical indication. Advancements in computing and information technology have led to many techniques of studying Alzheimers disease. Early identification and therapy are crucial for preventing Alzheimers disease, as early-onset dementia hits people before the age of 65, while late-onset dementia occurs after this age. According to the 2015 World Alzheimers disease Report, there are 46.8 million individuals worldwide suffering from dementia, with an anticipated 74.7 million more by 2030 and 131.5 million by 2050. Deep Learning has outperformed conventional Machine Learning techniques by identifying intricate structures in high-dimensional data. Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN), have achieved an accuracy of up to 96.0% for Alzheimers disease classification, and 84.2% for mild cognitive impairment (MCI) conversion prediction. There have been few literature surveys available on applying ML to predict dementia, lacking in congenital observations. However, this survey has focused on a specific data channel for dementia detection. This study evaluated Deep Learning algorithms for early Alzheimers disease detection, using openly accessible datasets, feature segmentation, and classification methods. This article also has identified research gaps and limits in detecting Alzheimers disease, which can inform future research.
- [143] arXiv:2502.05780 (replaced) [pdf, html, other]
-
Title: GOLD: Graph Out-of-Distribution Detection via Implicit Adversarial Latent GenerationComments: ICLR25Subjects: Machine Learning (cs.LG)
Despite graph neural networks' (GNNs) great success in modelling graph-structured data, out-of-distribution (OOD) test instances still pose a great challenge for current GNNs. One of the most effective techniques to detect OOD nodes is to expose the detector model with an additional OOD node-set, yet the extra OOD instances are often difficult to obtain in practice. Recent methods for image data address this problem using OOD data synthesis, typically relying on pre-trained generative models like Stable Diffusion. However, these approaches require vast amounts of additional data, as well as one-for-all pre-trained generative models, which are not available for graph data. Therefore, we propose the GOLD framework for graph OOD detection, an implicit adversarial learning pipeline with synthetic OOD exposure without pre-trained models. The implicit adversarial training process employs a novel alternating optimisation framework by training: (1) a latent generative model to regularly imitate the in-distribution (ID) embeddings from an evolving GNN, and (2) a GNN encoder and an OOD detector to accurately classify ID data while increasing the energy divergence between the ID embeddings and the generative model's synthetic embeddings. This novel approach implicitly transforms the synthetic embeddings into pseudo-OOD instances relative to the ID data, effectively simulating exposure to OOD scenarios without auxiliary data. Extensive OOD detection experiments are conducted on five benchmark graph datasets, verifying the superior performance of GOLD without using real OOD data compared with the state-of-the-art OOD exposure and non-exposure baselines.
- [144] arXiv:2502.06685 (replaced) [pdf, html, other]
-
Title: No Trick, No Treat: Pursuits and Challenges Towards Simulation-free Training of Neural SamplersJiajun He, Yuanqi Du, Francisco Vargas, Dinghuai Zhang, Shreyas Padhy, RuiKang OuYang, Carla Gomes, José Miguel Hernández-LobatoComments: 21 pages, 5 figures, 6 tablesSubjects: Machine Learning (cs.LG); Machine Learning (stat.ML)
We consider the sampling problem, where the aim is to draw samples from a distribution whose density is known only up to a normalization constant. Recent breakthroughs in generative modeling to approximate a high-dimensional data distribution have sparked significant interest in developing neural network-based methods for this challenging problem. However, neural samplers typically incur heavy computational overhead due to simulating trajectories during training. This motivates the pursuit of simulation-free training procedures of neural samplers. In this work, we propose an elegant modification to previous methods, which allows simulation-free training with the help of a time-dependent normalizing flow. However, it ultimately suffers from severe mode collapse. On closer inspection, we find that nearly all successful neural samplers rely on Langevin preconditioning to avoid mode collapsing. We systematically analyze several popular methods with various objective functions and demonstrate that, in the absence of Langevin preconditioning, most of them fail to adequately cover even a simple target. Finally, we draw attention to a strong baseline by combining the state-of-the-art MCMC method, Parallel Tempering (PT), with an additional generative model to shed light on future explorations of neural samplers.
- [145] arXiv:2502.07005 (replaced) [pdf, html, other]
-
Title: Geometry-aware RL for Manipulation of Varying Shapes and Deformable ObjectsComments: Accepted at ICLR 2025 (Oral)Subjects: Machine Learning (cs.LG); Robotics (cs.RO)
Manipulating objects with varying geometries and deformable objects is a major challenge in robotics. Tasks such as insertion with different objects or cloth hanging require precise control and effective modelling of complex dynamics. In this work, we frame this problem through the lens of a heterogeneous graph that comprises smaller sub-graphs, such as actuators and objects, accompanied by different edge types describing their interactions. This graph representation serves as a unified structure for both rigid and deformable objects tasks, and can be extended further to tasks comprising multiple actuators. To evaluate this setup, we present a novel and challenging reinforcement learning benchmark, including rigid insertion of diverse objects, as well as rope and cloth manipulation with multiple end-effectors. These tasks present a large search space, as both the initial and target configurations are uniformly sampled in 3D space. To address this issue, we propose a novel graph-based policy model, dubbed Heterogeneous Equivariant Policy (HEPi), utilizing $SE(3)$ equivariant message passing networks as the main backbone to exploit the geometric symmetry. In addition, by modeling explicit heterogeneity, HEPi can outperform Transformer-based and non-heterogeneous equivariant policies in terms of average returns, sample efficiency, and generalization to unseen objects. Our project page is available at this https URL.
- [146] arXiv:2502.07532 (replaced) [pdf, html, other]
-
Title: Diffusion-LAM: Probabilistic Limited Area Weather Forecasting with DiffusionComments: Accepted, camera ready versionSubjects: Machine Learning (cs.LG); Atmospheric and Oceanic Physics (physics.ao-ph)
Machine learning methods have been shown to be effective for weather forecasting, based on the speed and accuracy compared to traditional numerical models. While early efforts primarily concentrated on deterministic predictions, the field has increasingly shifted toward probabilistic forecasting to better capture the forecast uncertainty. Most machine learning-based models have been designed for global-scale predictions, with only limited work targeting regional or limited area forecasting, which allows more specialized and flexible modeling for specific locations. This work introduces Diffusion-LAM, a probabilistic limited area weather model leveraging conditional diffusion. By conditioning on boundary data from surrounding regions, our approach generates forecasts within a defined area. Experimental results on the MEPS limited area dataset demonstrate the potential of Diffusion-LAM to deliver accurate probabilistic forecasts, highlighting its promise for limited-area weather prediction.
- [147] arXiv:2502.10122 (replaced) [pdf, html, other]
-
Title: Modern Hopfield Networks with Continuous-Time MemoriesSubjects: Machine Learning (cs.LG)
Recent research has established a connection between modern Hopfield networks (HNs) and transformer attention heads, with guarantees of exponential storage capacity. However, these models still face challenges scaling storage efficiently. Inspired by psychological theories of continuous neural resource allocation in working memory, we propose an approach that compresses large discrete Hopfield memories into smaller, continuous-time memories. Leveraging continuous attention, our new energy function modifies the update rule of HNs, replacing the traditional softmax-based probability mass function with a probability density, over the continuous memory. This formulation aligns with modern perspectives on human executive function, offering a principled link between attractor dynamics in working memory and resource-efficient memory allocation. Our framework maintains competitive performance with HNs while leveraging a compressed memory, reducing computational costs across synthetic and video datasets.
- [148] arXiv:2503.00961 (replaced) [pdf, html, other]
-
Title: CAGN-GAT Fusion: A Hybrid Contrastive Attentive Graph Neural Network for Network Intrusion DetectionComments: Accepted in 38th International Conference on Industrial, Engineering & Other Applications of Applied Intelligent Systems (IEA/AIE 2025), Kitakyushu, Japan, Jul 2025Subjects: Machine Learning (cs.LG)
Cybersecurity threats are growing, making network intrusion detection essential. Traditional machine learning models remain effective in resource-limited environments due to their efficiency, requiring fewer parameters and less computational time. However, handling short and highly imbalanced datasets remains challenging. In this study, we propose the fusion of a Contrastive Attentive Graph Network and Graph Attention Network (CAGN-GAT Fusion) and benchmark it against 15 other models, including both Graph Neural Networks (GNNs) and traditional ML models. Our evaluation is conducted on four benchmark datasets (KDD-CUP-1999, NSL-KDD, UNSW-NB15, and CICIDS2017) using a short and proportionally imbalanced dataset with a constant size of 5000 samples to ensure fairness in comparison. Results show that CAGN-GAT Fusion demonstrates stable and competitive accuracy, recall, and F1-score, even though it does not achieve the highest performance in every dataset. Our analysis also highlights the impact of adaptive graph construction techniques, including small changes in connections (edge perturbation) and selective hiding of features (feature masking), improving detection performance. The findings confirm that GNNs, particularly CAGN-GAT Fusion, are robust and computationally efficient, making them well-suited for resource-constrained environments. Future work will explore GraphSAGE layers and multiview graph construction techniques to further enhance adaptability and detection accuracy.
- [149] arXiv:2503.08200 (replaced) [pdf, html, other]
-
Title: Route Sparse Autoencoder to Interpret Large Language ModelsSubjects: Machine Learning (cs.LG)
Mechanistic interpretability of large language models (LLMs) aims to uncover the internal processes of information propagation and reasoning. Sparse autoencoders (SAEs) have demonstrated promise in this domain by extracting interpretable and monosemantic features. However, prior works primarily focus on feature extraction from a single layer, failing to effectively capture activations that span multiple layers. In this paper, we introduce Route Sparse Autoencoder (RouteSAE), a new framework that integrates a routing mechanism with a shared SAE to efficiently extract features from multiple layers. It dynamically assigns weights to activations from different layers, incurring minimal parameter overhead while achieving high interpretability and flexibility for targeted feature manipulation. We evaluate RouteSAE through extensive experiments on Llama-3.2-1B-Instruct. Specifically, under the same sparsity constraint of 64, RouteSAE extracts 22.5% more features than baseline SAEs while achieving a 22.3% higher interpretability score. These results underscore the potential of RouteSAE as a scalable and effective method for LLM interpretability, with applications in feature discovery and model intervention. Our codes are available at this https URL.
- [150] arXiv:2503.13366 (replaced) [pdf, html, other]
-
Title: Optimal Bounds for Adversarial Constrained Online Convex OptimizationSubjects: Machine Learning (cs.LG); Data Structures and Algorithms (cs.DS); Optimization and Control (math.OC); Machine Learning (stat.ML)
Constrained Online Convex Optimization (COCO) can be seen as a generalization of the standard Online Convex Optimization (OCO) framework. At each round, a cost function and constraint function are revealed after a learner chooses an action. The goal is to minimize both the regret and cumulative constraint violation (CCV) against an adaptive adversary. We show for the first time that is possible to obtain the optimal $O(\sqrt{T})$ bound on both regret and CCV, improving the best known bounds of $O \left( \sqrt{T} \right)$ and $\tilde{O} \left( \sqrt{T} \right)$ for the regret and CCV, respectively. Based on a new surrogate loss function enforcing a minimum penalty on the constraint function, we demonstrate that both the Follow-the-Regularized-Leader and the Online Gradient Descent achieve the optimal bounds.
- [151] arXiv:2503.13911 (replaced) [pdf, html, other]
-
Title: Incorporating Attributes and Multi-Scale Structures for Heterogeneous Graph Contrastive LearningSubjects: Machine Learning (cs.LG); Social and Information Networks (cs.SI)
Heterogeneous graphs (HGs) are composed of multiple types of nodes and edges, making it more effective in capturing the complex relational structures inherent in the real world. However, in real-world scenarios, labeled data is often difficult to obtain, which limits the applicability of semi-supervised approaches. Self-supervised learning aims to enable models to automatically learn useful features from data, effectively addressing the challenge of limited labeling data. In this paper, we propose a novel contrastive learning framework for heterogeneous graphs (ASHGCL), which incorporates three distinct views, each focusing on node attributes, high-order and low-order structural information, respectively, to effectively capture attribute information, high-order structures, and low-order structures for node representation learning. Furthermore, we introduce an attribute-enhanced positive sample selection strategy that combines both structural information and attribute information, effectively addressing the issue of sampling bias. Extensive experiments on four real-world datasets show that ASHGCL outperforms state-of-the-art unsupervised baselines and even surpasses some supervised benchmarks.
- [152] arXiv:2503.22418 (replaced) [pdf, html, other]
-
Title: Robustness quantification: a new method for assessing the reliability of the predictions of a classifierSubjects: Machine Learning (cs.LG); Probability (math.PR)
Based on existing ideas in the field of imprecise probabilities, we present a new approach for assessing the reliability of the individual predictions of a generative probabilistic classifier. We call this approach robustness quantification, compare it to uncertainty quantification, and demonstrate that it continues to work well even for classifiers that are learned from small training sets that are sampled from a shifted distribution.
- [153] arXiv:2503.23820 (replaced) [pdf, html, other]
-
Title: When Counterfactual Reasoning Fails: Chaos and Real-World ComplexitySubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Counterfactual reasoning, a cornerstone of human cognition and decision-making, is often seen as the 'holy grail' of causal learning, with applications ranging from interpreting machine learning models to promoting algorithmic fairness. While counterfactual reasoning has been extensively studied in contexts where the underlying causal model is well-defined, real-world causal modeling is often hindered by model and parameter uncertainty, observational noise, and chaotic behavior. The reliability of counterfactual analysis in such settings remains largely unexplored. In this work, we investigate the limitations of counterfactual reasoning within the framework of Structural Causal Models. Specifically, we empirically investigate \emph{counterfactual sequence estimation} and highlight cases where it becomes increasingly unreliable. We find that realistic assumptions, such as low degrees of model uncertainty or chaotic dynamics, can result in counterintuitive outcomes, including dramatic deviations between predicted and true counterfactual trajectories. This work urges caution when applying counterfactual reasoning in settings characterized by chaos and uncertainty. Furthermore, it raises the question of whether certain systems may pose fundamental limitations on the ability to answer counterfactual questions about their behavior.
- [154] arXiv:2503.24305 (replaced) [pdf, html, other]
-
Title: Evaluating machine learning models for predicting pesticides toxicity to honey beesSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Small molecules play a critical role in the biomedical, environmental, and agrochemical domains, each with distinct physicochemical requirements and success criteria. Although biomedical research benefits from extensive datasets and established benchmarks, agrochemical data remain scarce, particularly with respect to species-specific toxicity. This work focuses on ApisTox, the most comprehensive dataset of experimentally validated chemical toxicity to the honey bee (Apis mellifera), an ecologically vital pollinator. We evaluate ApisTox using a diverse suite of machine learning approaches, including molecular fingerprints, graph kernels, and graph neural networks, as well as pretrained models. Comparative analysis with medicinal datasets from the MoleculeNet benchmark reveals that ApisTox represents a distinct chemical space. Performance degradation on non-medicinal datasets, such as ApisTox, demonstrates their limited generalizability of current state-of-the-art algorithms trained solely on biomedical data. Our study highlights the need for more diverse datasets and for targeted model development geared toward the agrochemical domain.
- [155] arXiv:2504.03122 (replaced) [pdf, html, other]
-
Title: From Observation to Orientation: an Adaptive Integer Programming Approach to Intervention DesignSubjects: Machine Learning (cs.LG); Machine Learning (stat.ML)
Using both observational and experimental data, a causal discovery process can identify the causal relationships between variables. A unique adaptive intervention design paradigm is presented in this work, where causal directed acyclic graphs (DAGs) are for effectively recovered with practical budgetary considerations. In order to choose treatments that optimize information gain under these considerations, an iterative integer programming (IP) approach is proposed, which drastically reduces the number of experiments required. Simulations over a broad range of graph sizes and edge densities are used to assess the effectiveness of the suggested approach. Results show that the proposed adaptive IP approach achieves full causal graph recovery with fewer intervention iterations and variable manipulations than random intervention baselines, and it is also flexible enough to accommodate a variety of practical constraints.
- [156] arXiv:2504.03783 (replaced) [pdf, html, other]
-
Title: FAST: Federated Active Learning with Foundation Models for Communication-efficient Sampling and TrainingSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Distributed, Parallel, and Cluster Computing (cs.DC)
Federated Active Learning (FAL) has emerged as a promising framework to leverage large quantities of unlabeled data across distributed clients while preserving data privacy. However, real-world deployments remain limited by high annotation costs and communication-intensive sampling processes, particularly in a cross-silo setting, when clients possess substantial local datasets. This paper addresses the crucial question: What is the best practice to reduce communication costs in human-in-the-loop learning with minimal annotator effort? Existing FAL methods typically rely on iterative annotation processes that separate active sampling from federated updates, leading to multiple rounds of expensive communication and annotation. In response, we introduce FAST, a two-pass FAL framework that harnesses foundation models for weak labeling in a preliminary pass, followed by a refinement pass focused exclusively on the most uncertain samples. By leveraging representation knowledge from foundation models and integrating refinement steps into a streamlined workflow, FAST substantially reduces the overhead incurred by iterative active sampling. Extensive experiments on diverse medical and natural image benchmarks demonstrate that FAST outperforms existing FAL methods by an average of 4.36% while reducing communication rounds eightfold under a limited 5% labeling budget.
- [157] arXiv:2504.05586 (replaced) [pdf, html, other]
-
Title: Finding Fantastic Experts in MoEs: A Unified Study for Expert Dropping Strategies and ObservationsAjay Jaiswal, Jianyu Wang, Yixiao Li, Pingzhi Li, Tianlong Chen, Zhangyang Wang, Chong Wang, Ruoming Pang, Xianzhi DuSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Sparsely activated Mixture-of-Experts (SMoE) has shown promise in scaling up the learning capacity of neural networks. However, vanilla SMoEs have issues such as expert redundancy and heavy memory requirements, making them inefficient and non-scalable, especially for resource-constrained scenarios. Expert-level sparsification of SMoEs involves pruning the least important experts to address these limitations. In this work, we aim to address three questions: (1) What is the best recipe to identify the least knowledgeable subset of experts that can be dropped with minimal impact on performance? (2) How should we perform expert dropping (one-shot or iterative), and what correction measures can we undertake to minimize its drastic impact on SMoE subnetwork capabilities? (3) What capabilities of full-SMoEs are severely impacted by the removal of the least dominant experts, and how can we recover them? Firstly, we propose MoE Experts Compression Suite (MC-Suite), which is a collection of some previously explored and multiple novel recipes to provide a comprehensive benchmark for estimating expert importance from diverse perspectives, as well as unveil numerous valuable insights for SMoE experts. Secondly, unlike prior works with a one-shot expert pruning approach, we explore the benefits of iterative pruning with the re-estimation of the MC-Suite criterion. Moreover, we introduce the benefits of task-agnostic fine-tuning as a correction mechanism during iterative expert dropping, which we term MoE Lottery Subnetworks. Lastly, we present an experimentally validated conjecture that, during expert dropping, SMoEs' instruction-following capabilities are predominantly hurt, which can be restored to a robust level subject to external augmentation of instruction-following capabilities using k-shot examples and supervised fine-tuning.
- [158] arXiv:2504.05761 (replaced) [pdf, html, other]
-
Title: AiGAS-dEVL-RC: An Adaptive Growing Neural Gas Model for Recurrently Drifting Unsupervised Data StreamsComments: Copyright 2025 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other worksSubjects: Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)
Concept drift and extreme verification latency pose significant challenges in data stream learning, particularly when dealing with recurring concept changes in dynamic environments. This work introduces a novel method based on the Growing Neural Gas (GNG) algorithm, designed to effectively handle abrupt recurrent drifts while adapting to incrementally evolving data distributions (incremental drifts). Leveraging the self-organizing and topological adaptability of GNG, the proposed approach maintains a compact yet informative memory structure, allowing it to efficiently store and retrieve knowledge of past or recurring concepts, even under conditions of delayed or sparse stream supervision. Our experiments highlight the superiority of our approach over existing data stream learning methods designed to cope with incremental non-stationarities and verification latency, demonstrating its ability to quickly adapt to new drifts, robustly manage recurring patterns, and maintain high predictive accuracy with a minimal memory footprint. Unlike other techniques that fail to leverage recurring knowledge, our proposed approach is proven to be a robust and efficient online learning solution for unsupervised drifting data flows.
- [159] arXiv:2504.06212 (replaced) [pdf, html, other]
-
Title: NNN: Next-Generation Neural Networks for Marketing Mix ModelingSubjects: Machine Learning (cs.LG); Applications (stat.AP)
We present NNN, a Transformer-based neural network approach to Marketing Mix Modeling (MMM) designed to address key limitations of traditional methods. Unlike conventional MMMs which rely on scalar inputs and parametric decay functions, NNN uses rich embeddings to capture both quantitative and qualitative aspects of marketing and organic channels (e.g., search queries, ad creatives). This, combined with its attention mechanism, enables NNN to model complex interactions, capture long-term effects, and potentially improve sales attribution accuracy. We show that L1 regularization permits the use of such expressive models in typical data-constrained settings. Evaluating NNN on simulated and real-world data demonstrates its efficacy, particularly through considerable improvement in predictive power. Beyond attribution, NNN provides valuable, complementary insights through model probing, such as evaluating keyword or creative effectiveness, enhancing model interpretability.
- [160] arXiv:2504.06265 (replaced) [pdf, html, other]
-
Title: GOLLuM: Gaussian Process Optimized LLMs -- Reframing LLM Finetuning through Bayesian OptimizationSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Large Language Models (LLMs) can encode complex relationships in their latent spaces, yet harnessing them for optimization under uncertainty remains challenging. We address this gap with a novel architecture that reframes LLM finetuning as Gaussian process (GP) marginal likelihood optimization via deep kernel methods. We introduce LLM-based deep kernels, jointly optimized with GPs to preserve the benefits of both - LLMs to provide a rich and flexible input space for Bayesian optimization and - GPs to model this space with predictive uncertainty for more efficient sampling. Applied to Buchwald-Hartwig reaction optimization, our method nearly doubles the discovery rate of high-performing reactions compared to static LLM embeddings (from 24% to 43% coverage of the top 5% reactions in just 50 optimization iterations). We also observe a 14% improvement over domain-specific representations without requiring specialized features. Extensive empirical evaluation across 19 benchmarks - ranging from general chemistry to reaction and molecular property optimization - demonstrates our method's robustness, generality, and consistent improvements across: (1) tasks, (2) LLM architectures (encoder, decoder, encoder-decoder), (3) pretraining domains (chemistry-related or general-purpose) and (4) hyperparameter settings (tuned once on a single dataset). Finally, we explain these improvements: joint LLM-GP optimization through marginal likelihood implicitly performs contrastive learning, aligning representations to produce (1) better-structured embedding spaces, (2) improved uncertainty calibration, and (3) more efficient sampling - without requiring any external loss. This work provides both practical advances in sample-efficient optimization and insights into what makes effective Bayesian optimization.
- [161] arXiv:2504.06611 (replaced) [pdf, html, other]
-
Title: Wanting to be UnderstoodSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
This paper explores an intrinsic motivation for mutual awareness, hypothesizing that humans possess a fundamental drive to understand and to be understood even in the absence of extrinsic rewards. Through simulations of the perceptual crossing paradigm, we explore the effect of various internal reward functions in reinforcement learning agents. The drive to understand is implemented as an active inference type artificial curiosity reward, whereas the drive to be understood is implemented through intrinsic rewards for imitation, influence/impressionability, and sub-reaction time anticipation of the other. Results indicate that while artificial curiosity alone does not lead to a preference for social interaction, rewards emphasizing reciprocal understanding successfully drive agents to prioritize interaction. We demonstrate that this intrinsic motivation can facilitate cooperation in tasks where only one agent receives extrinsic reward for the behaviour of the other.
- [162] arXiv:2504.06643 (replaced) [pdf, html, other]
-
Title: AMAD: AutoMasked Attention for Unsupervised Multivariate Time Series Anomaly DetectionComments: fix img issuesSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Unsupervised multivariate time series anomaly detection (UMTSAD) plays a critical role in various domains, including finance, networks, and sensor systems. In recent years, due to the outstanding performance of deep learning in general sequential tasks, many models have been specialized for deep UMTSAD tasks and have achieved impressive results, particularly those based on the Transformer and self-attention mechanisms. However, the sequence anomaly association assumptions underlying these models are often limited to specific predefined patterns and scenarios, such as concentrated or peak anomaly patterns. These limitations hinder their ability to generalize to diverse anomaly situations, especially where the lack of labels poses significant challenges. To address these issues, we propose AMAD, which integrates \textbf{A}uto\textbf{M}asked Attention for UMTS\textbf{AD} scenarios. AMAD introduces a novel structure based on the AutoMask mechanism and an attention mixup module, forming a simple yet generalized anomaly association representation framework. This framework is further enhanced by a Max-Min training strategy and a Local-Global contrastive learning approach. By combining multi-scale feature extraction with automatic relative association modeling, AMAD provides a robust and adaptable solution to UMTSAD challenges. Extensive experimental results demonstrate that the proposed model achieving competitive performance results compared to SOTA benchmarks across a variety of datasets.
- [163] arXiv:2210.13300 (replaced) [pdf, html, other]
-
Title: Designing Universal Causal Deep Learning Models: The Case of Infinite-Dimensional Dynamical Systems from Stochastic AnalysisSubjects: Dynamical Systems (math.DS); Machine Learning (cs.LG); Computational Finance (q-fin.CP)
Several non-linear operators in stochastic analysis, such as solution maps to stochastic differential equations, depend on a temporal structure which is not leveraged by contemporary neural operators designed to approximate general maps between Banach space. This paper therefore proposes an operator learning solution to this open problem by introducing a deep learning model-design framework that takes suitable infinite-dimensional linear metric spaces, e.g. Banach spaces, as inputs and returns a universal \textit{sequential} deep learning model adapted to these linear geometries specialized for the approximation of operators encoding a temporal structure. We call these models \textit{Causal Neural Operators}. Our main result states that the models produced by our framework can uniformly approximate on compact sets and across arbitrarily finite-time horizons Hölder or smooth trace class operators, which causally map sequences between given linear metric spaces. Our analysis uncovers new quantitative relationships on the latent state-space dimension of Causal Neural Operators, which even have new implications for (classical) finite-dimensional Recurrent Neural Networks. In addition, our guarantees for recurrent neural networks are tighter than the available results inherited from feedforward neural networks when approximating dynamical systems between finite-dimensional spaces.
- [164] arXiv:2302.07425 (replaced) [pdf, html, other]
-
Title: Bandit Social Learning: Exploration under Myopic BehaviorComments: Extended version of NeurIPS 2023 paper titled "Bandit Social Learning under Myopic Behavior"Subjects: Computer Science and Game Theory (cs.GT); Data Structures and Algorithms (cs.DS); Machine Learning (cs.LG)
We study social learning dynamics motivated by reviews on online platforms. The agents collectively follow a simple multi-armed bandit protocol, but each agent acts myopically, without regards to exploration. We allow the greedy (exploitation-only) algorithm, as well as a wide range of behavioral biases. Specifically, we allow myopic behaviors that are consistent with (parameterized) confidence intervals for the arms' expected rewards. We derive stark learning failures for any such behavior, and provide matching positive results. The learning-failure results extend to Bayesian agents and Bayesian bandit environments.
In particular, we obtain general, quantitatively strong results on failure of the greedy bandit algorithm, both for ``frequentist" and ``Bayesian" versions. Failure results known previously are quantitatively weak, and either trivial or very specialized. Thus, we provide a theoretical foundation for designing non-trivial bandit algorithms, \ie algorithms that intentionally explore, which has been missing from the literature.
Our general behavioral model can be interpreted as agents' optimism or pessimism. The matching positive results entail a maximal allowed amount of optimism. Moreover, we find that no amount of pessimism helps against the learning failures, whereas even a small-but-constant fraction of extreme optimists avoids the failures and leads to near-optimal regret rates. - [165] arXiv:2307.02284 (replaced) [pdf, html, other]
-
Title: Universal Scaling Laws of Absorbing Phase Transitions in Artificial Deep Neural NetworksComments: 15 pages, 5 figures; added ReLU finite-size scaling results, revised texts for claritySubjects: Machine Learning (stat.ML); Disordered Systems and Neural Networks (cond-mat.dis-nn); Statistical Mechanics (cond-mat.stat-mech); Machine Learning (cs.LG)
We demonstrate that conventional artificial deep neural networks operating near the phase boundary of the signal propagation dynamics, also known as the edge of chaos, exhibit universal scaling laws of absorbing phase transitions in non-equilibrium statistical mechanics. We exploit the fully deterministic nature of the propagation dynamics to elucidate an analogy between a signal collapse in the neural networks and an absorbing state (a state that the system can enter but cannot escape from). Our numerical results indicate that the multilayer perceptrons and the convolutional neural networks belong to the mean-field and the directed percolation universality classes, respectively. Also, the finite-size scaling is successfully applied, suggesting a potential connection to the depth-width trade-off in deep learning. Furthermore, our analysis of the training dynamics under the gradient descent reveals that hyperparameter tuning to the phase boundary is necessary but insufficient for achieving optimal generalization in deep networks. Remarkably, nonuniversal metric factors associated with the scaling laws are shown to play a significant role in concretizing the above observations. These findings highlight the usefulness of the notion of criticality for analyzing the behavior of artificial deep neural networks and offer new insights toward a unified understanding of the essential relationship between criticality and intelligence.
- [166] arXiv:2307.08813 (replaced) [pdf, html, other]
-
Title: Comparative Performance Evaluation of Large Language Models for Extracting Molecular Interactions and Pathway KnowledgeSubjects: Computation and Language (cs.CL); Machine Learning (cs.LG)
Background Identification of the interactions and regulatory relations between biomolecules play pivotal roles in understanding complex biological systems and the mechanisms underlying diverse biological functions. However, the collection of such molecular interactions has heavily relied on expert curation in the past, making it labor-intensive and time-consuming. To mitigate these challenges, we propose leveraging the capabilities of large language models (LLMs) to automate genome-scale extraction of this crucial knowledge.
Results In this study, we investigate the efficacy of various LLMs in addressing biological tasks, such as the recognition of protein interactions, identification of genes linked to pathways affected by low-dose radiation, and the delineation of gene regulatory relationships. Overall, the larger models exhibited superior performance, indicating their potential for specific tasks that involve the extraction of complex interactions among genes and proteins. Although these models possessed detailed information for distinct gene and protein groups, they faced challenges in identifying groups with diverse functions and in recognizing highly correlated gene regulatory relationships.
Conclusions By conducting a comprehensive assessment of the state-of-the-art models using well-established molecular interaction and pathway databases, our study reveals that LLMs can identify genes/proteins associated with pathways of interest and predict their interactions to a certain extent. Furthermore, these models can provide important insights, marking a noteworthy stride toward advancing our understanding of biological systems through AI-assisted knowledge discovery. - [167] arXiv:2310.05273 (replaced) [pdf, html, other]
-
Title: Physics-tailored machine learning reveals unexpected physics in dusty plasmasComments: 15 pages, 4 Figures, 2 Supplemental Figures, 8 Supplemental VideosSubjects: Plasma Physics (physics.plasm-ph); Machine Learning (cs.LG); Computational Physics (physics.comp-ph)
Dusty plasma is a mixture of ions, electrons, and macroscopic charged particles that is commonly found in space and planetary environments. The particles interact through Coulomb forces mediated by the surrounding plasma, and as a result, the effective forces between particles can be non-conservative and non-reciprocal. Machine learning (ML) models are a promising route to learn these complex forces, yet their structure should match the underlying physical constraints to provide useful insight. Here we demonstrate and experimentally validate an ML approach that incorporates physical intuition to infer force laws in a laboratory dusty plasma. Trained on 3D particle trajectories, the model accounts for inherent symmetries, non-identical particles, and learns the effective non-reciprocal forces between particles with exquisite accuracy (R^2>0.99). We validate the model by inferring particle masses in two independent yet consistent ways. The model's accuracy enables precise measurements of particle charge and screening length, discovering large deviations from common theoretical assumptions. Our ability to identify new physics from experimental data demonstrates how ML-powered approaches can guide new routes of scientific discovery in many-body systems. Furthermore, we anticipate our ML approach to be a starting point for inferring laws from dynamics in a wide range of many-body systems, from colloids to living organisms.
- [168] arXiv:2310.07891 (replaced) [pdf, other]
-
Title: A Theory of Non-Linear Feature Learning with One Gradient Step in Two-Layer Neural NetworksSubjects: Machine Learning (stat.ML); Machine Learning (cs.LG)
Feature learning is thought to be one of the fundamental reasons for the success of deep neural networks. It is rigorously known that in two-layer fully-connected neural networks under certain conditions, one step of gradient descent on the first layer can lead to feature learning; characterized by the appearance of a separated rank-one component -- spike -- in the spectrum of the feature matrix. However, with a constant gradient descent step size, this spike only carries information from the linear component of the target function and therefore learning non-linear components is impossible. We show that with a learning rate that grows with the sample size, such training in fact introduces multiple rank-one components, each corresponding to a specific polynomial feature. We further prove that the limiting large-dimensional and large sample training and test errors of the updated neural networks are fully characterized by these spikes. By precisely analyzing the improvement in the training and test errors, we demonstrate that these non-linear features can enhance learning.
- [169] arXiv:2310.12806 (replaced) [pdf, html, other]
-
Title: DCSI -- An improved measure of cluster separability based on separation and connectednessSubjects: Machine Learning (stat.ML); Machine Learning (cs.LG)
Whether class labels in a given data set correspond to meaningful clusters is crucial for the evaluation of clustering algorithms using real-world data sets. This property can be quantified by separability measures. The central aspects of separability for density-based clustering are between-class separation and within-class connectedness, and neither classification-based complexity measures nor cluster validity indices (CVIs) adequately incorporate them. A newly developed measure (density cluster separability index, DCSI) aims to quantify these two characteristics and can also be used as a CVI. Extensive experiments on synthetic data indicate that DCSI correlates strongly with the performance of DBSCAN measured via the adjusted Rand index (ARI) but lacks robustness when it comes to multi-class data sets with overlapping classes that are ill-suited for density-based hard clustering. Detailed evaluation on frequently used real-world data sets shows that DCSI can correctly identify touching or overlapping classes that do not correspond to meaningful density-based clusters.
- [170] arXiv:2404.08335 (replaced) [pdf, html, other]
-
Title: Toward a Theory of Tokenization in LLMsComments: 60 pages, 11 figures. This work was published at NeurIPS 2024 with a different title, "An Analysis of Tokenization: Transformers under Markov data"Subjects: Computation and Language (cs.CL); Machine Learning (cs.LG)
While there has been a large body of research attempting to circumvent tokenization for language modeling (Clark et al., 2022; Xue et al., 2022), the current consensus is that it is a necessary initial step for designing state-of-the-art performant language models. In this paper, we investigate tokenization from a theoretical point of view by studying the behavior of transformers on simple data generating processes. When trained on data drawn from certain simple $k^{\text{th}}$-order Markov processes for $k > 1$, transformers exhibit a surprising phenomenon - in the absence of tokenization, they empirically fail to learn the right distribution and predict characters according to a unigram model (Makkuva et al., 2024). With the addition of tokenization, however, we empirically observe that transformers break through this barrier and are able to model the probabilities of sequences drawn from the source near-optimally, achieving small cross-entropy loss. With this observation as starting point, we study the end-to-end cross-entropy loss achieved by transformers with and without tokenization. With the appropriate tokenization, we show that even the simplest unigram models (over tokens) learnt by transformers are able to model the probability of sequences drawn from $k^{\text{th}}$-order Markov sources near optimally. Our analysis provides a justification for the use of tokenization in practice through studying the behavior of transformers on Markovian data.
- [171] arXiv:2405.18560 (replaced) [pdf, html, other]
-
Title: Potential Field Based Deep Metric LearningComments: Accepted to CVPR 2025Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR); Machine Learning (cs.LG); Image and Video Processing (eess.IV)
Deep metric learning (DML) involves training a network to learn a semantically meaningful representation space. Many current approaches mine n-tuples of examples and model interactions within each tuplets. We present a novel, compositional DML model that instead of in tuples, represents the influence of each example (embedding) by a continuous potential field, and superposes the fields to obtain their combined global potential field. We use attractive/repulsive potential fields to represent interactions among embeddings from images of the same/different classes. Contrary to typical learning methods, where mutual influence of samples is proportional to their distance, we enforce reduction in such influence with distance, leading to a decaying field. We show that such decay helps improve performance on real world datasets with large intra-class variations and label noise. Like other proxy-based methods, we also use proxies to succinctly represent sub-populations of examples. We evaluate our method on three standard DML benchmarks- Cars-196, CUB-200-2011, and SOP datasets where it outperforms state-of-the-art baselines.
- [172] arXiv:2406.07409 (replaced) [pdf, html, other]
-
Title: Accelerating Ill-conditioned Hankel Matrix Recovery via Structured Newton-like DescentSubjects: Machine Learning (stat.ML); Information Theory (cs.IT); Machine Learning (cs.LG); Signal Processing (eess.SP); Optimization and Control (math.OC)
This paper studies the robust Hankel recovery problem, which simultaneously removes the sparse outliers and fulfills missing entries from the partial observation. We propose a novel non-convex algorithm, coined Hankel Structured Newton-Like Descent (HSNLD), to tackle the robust Hankel recovery problem. HSNLD is highly efficient with linear convergence, and its convergence rate is independent of the condition number of the underlying Hankel matrix. The recovery guarantee has been established under some mild conditions. Numerical experiments on both synthetic and real datasets show the superior performance of HSNLD against state-of-the-art algorithms.
- [173] arXiv:2406.12123 (replaced) [pdf, html, other]
-
Title: ChatEMG: Synthetic Data Generation to Control a Robotic Hand Orthosis for StrokeJingxi Xu, Runsheng Wang, Siqi Shang, Ava Chen, Lauren Winterbottom, To-Liang Hsu, Wenxi Chen, Khondoker Ahmed, Pedro Leandro La Rotta, Xinyue Zhu, Dawn M. Nilsen, Joel Stein, Matei CiocarlieComments: 8 pages; accepted to RA-L in November 2024; published at RA-L in February 2025Subjects: Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Intent inferral on a hand orthosis for stroke patients is challenging due to the difficulty of data collection. Additionally, EMG signals exhibit significant variations across different conditions, sessions, and subjects, making it hard for classifiers to generalize. Traditional approaches require a large labeled dataset from the new condition, session, or subject to train intent classifiers; however, this data collection process is burdensome and time-consuming. In this paper, we propose ChatEMG, an autoregressive generative model that can generate synthetic EMG signals conditioned on prompts (i.e., a given sequence of EMG signals). ChatEMG enables us to collect only a small dataset from the new condition, session, or subject and expand it with synthetic samples conditioned on prompts from this new context. ChatEMG leverages a vast repository of previous data via generative training while still remaining context-specific via prompting. Our experiments show that these synthetic samples are classifier-agnostic and can improve intent inferral accuracy for different types of classifiers. We demonstrate that our complete approach can be integrated into a single patient session, including the use of the classifier for functional orthosis-assisted tasks. To the best of our knowledge, this is the first time an intent classifier trained partially on synthetic data has been deployed for functional control of an orthosis by a stroke survivor. Videos, source code, and additional information can be found at this https URL.
- [174] arXiv:2407.00085 (replaced) [pdf, html, other]
-
Title: Compressing Search with Language ModelsComments: Fixed an error in Appendix A. We were referencing |S|-length variables when they should have been D-lengthSubjects: Information Retrieval (cs.IR); Machine Learning (cs.LG)
Millions of people turn to Google Search each day for information on things as diverse as new cars or flu symptoms. The terms that they enter contain valuable information on their daily intent and activities, but the information in these search terms has been difficult to fully leverage. User-defined categorical filters have been the most common way to shrink the dimensionality of search data to a tractable size for analysis and modeling. In this paper we present a new approach to reducing the dimensionality of search data while retaining much of the information in the individual terms without user-defined rules. Our contributions are two-fold: 1) we introduce SLaM Compression, a way to quantify search terms using pre-trained language models and create a representation of search data that has low dimensionality, is memory efficient, and effectively acts as a summary of search, and 2) we present CoSMo, a Constrained Search Model for estimating real world events using only search data. We demonstrate the efficacy of our contributions by estimating with high accuracy U.S. automobile sales and U.S. flu rates using only Google Search data.
- [175] arXiv:2407.09722 (replaced) [pdf, html, other]
-
Title: Optimized Multi-Token Joint Decoding with Auxiliary Model for LLM InferenceJournal-ref: ICLR 2025Subjects: Computation and Language (cs.CL); Machine Learning (cs.LG)
Large language models (LLMs) have achieved remarkable success across diverse tasks, yet their inference processes are hindered by substantial time and energy demands due to single-token generation at each decoding step. While previous methods such as speculative decoding mitigate these inefficiencies by producing multiple tokens per step, each token is still generated by its single-token distribution, thereby enhancing speed without improving effectiveness. In contrast, our work simultaneously enhances inference speed and improves the output effectiveness. We consider multi-token joint decoding (MTJD), which generates multiple tokens from their joint distribution at each iteration, theoretically reducing perplexity and enhancing task performance. However, MTJD suffers from the high cost of sampling from the joint distribution of multiple tokens. Inspired by speculative decoding, we introduce multi-token assisted decoding (MTAD), a novel framework designed to accelerate MTJD. MTAD leverages a smaller auxiliary model to approximate the joint distribution of a larger model, incorporating a verification mechanism that not only ensures the accuracy of this approximation, but also improves the decoding efficiency over conventional speculative decoding. Theoretically, we demonstrate that MTAD closely approximates exact MTJD with bounded error. Empirical evaluations using Llama-2 and OPT models ranging from 13B to 70B parameters across various tasks reveal that MTAD reduces perplexity by 21.2% and improves downstream performance compared to standard single-token sampling. Furthermore, MTAD achieves a 1.42x speed-up and consumes 1.54x less energy than conventional speculative decoding methods. These results highlight MTAD's ability to make multi-token joint decoding both effective and efficient, promoting more sustainable and high-performance deployment of LLMs.
- [176] arXiv:2408.07644 (replaced) [pdf, other]
-
Title: SigmaRL: A Sample-Efficient and Generalizable Multi-Agent Reinforcement Learning Framework for Motion PlanningComments: Accepted for presentation at the IEEE International Conference on Intelligent Transportation Systems (ITSC) 2024Subjects: Robotics (cs.RO); Machine Learning (cs.LG); Multiagent Systems (cs.MA); Systems and Control (eess.SY)
This paper introduces an open-source, decentralized framework named SigmaRL, designed to enhance both sample efficiency and generalization of multi-agent Reinforcement Learning (RL) for motion planning of connected and automated vehicles. Most RL agents exhibit a limited capacity to generalize, often focusing narrowly on specific scenarios, and are usually evaluated in similar or even the same scenarios seen during training. Various methods have been proposed to address these challenges, including experience replay and regularization. However, how observation design in RL affects sample efficiency and generalization remains an under-explored area. We address this gap by proposing five strategies to design information-dense observations, focusing on general features that are applicable to most traffic scenarios. We train our RL agents using these strategies on an intersection and evaluate their generalization through numerical experiments across completely unseen traffic scenarios, including a new intersection, an on-ramp, and a roundabout. Incorporating these information-dense observations reduces training times to under one hour on a single CPU, and the evaluation results reveal that our RL agents can effectively zero-shot generalize. Code: this http URL
- [177] arXiv:2410.03041 (replaced) [pdf, html, other]
-
Title: Minmax Trend Filtering: Generalizations of Total Variation Denoising via a Local Minmax/Maxmin FormulaSubjects: Statistics Theory (math.ST); Machine Learning (cs.LG)
Total Variation Denoising (TVD) is a fundamental denoising and smoothing method. In this article, we identify a new local minmax/maxmin formula producing two estimators which sandwich the univariate TVD estimator at every point. Operationally, this formula gives a local definition of TVD as a minmax/maxmin of a simple function of local averages. Moreover we find that this minmax/maxmin formula is generalizeable and can be used to define other TVD like estimators. In this article we propose and study higher order polynomial versions of TVD which are defined pointwise lying between minmax and maxmin optimizations of penalized local polynomial regressions over intervals of different scales. These appear to be new nonparametric regression methods, different from usual Trend Filtering and any other existing method in the nonparametric regression toolbox. We call these estimators Minmax Trend Filtering (MTF). We show how the proposed local definition of TVD/MTF estimator makes it tractable to bound pointwise estimation errors in terms of a local bias variance like trade-off. This type of local analysis of TVD/MTF is new and arguably simpler than existing analyses of TVD/Trend Filtering. In particular, apart from minimax rate optimality over bounded variation and piecewise polynomial classes, our pointwise estimation error bounds also enable us to derive local rates of convergence for (locally) Holder Smooth signals. These local rates offer a new pointwise explanation of local adaptivity of TVD/MTF instead of global (MSE) based justifications.
- [178] arXiv:2410.23912 (replaced) [pdf, html, other]
-
Title: RL-STaR: Theoretical Analysis of Reinforcement Learning Frameworks for Self-Taught ReasonerJournal-ref: ICLR 2025 Workshop on Reasoning and Planning for Large Language ModelsSubjects: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
The reasoning abilities of large language models (LLMs) have improved with chain-of-thought (CoT) prompting, allowing models to solve complex tasks stepwise. However, training CoT capabilities requires detailed reasoning data, which is often scarce. The self-taught reasoner (STaR) framework addresses this by using reinforcement learning to automatically generate reasoning steps, reducing reliance on human-labeled data. Although STaR and its variants have demonstrated empirical success, a theoretical foundation explaining these improvements is lacking. This work provides a theoretical framework for understanding the effectiveness of reinforcement learning on CoT reasoning and STaR. Our contributions are: (1) criteria for the quality of pre-trained models necessary to initiate effective reasoning improvement; (2) an analysis of policy improvement, showing why LLM reasoning improves iteratively with STaR; (3) conditions for convergence to an optimal reasoning policy; and (4) an examination of STaR's robustness, explaining how it can improve reasoning even when incorporating occasional incorrect steps; This framework aims to bridge empirical findings with theoretical insights, advancing reinforcement learning approaches for reasoning in LLMs.
- [179] arXiv:2411.16693 (replaced) [pdf, html, other]
-
Title: UQ of 2D Slab Burner DNS: Surrogates, Uncertainty Propagation, and Parameter CalibrationGeorgios Georgalis, Alejandro Becerra, Kenneth Budzinski, Matthew McGurn, Danial Faghihi, Paul E. DesJardin, Abani PatraSubjects: Computational Physics (physics.comp-ph); Machine Learning (cs.LG)
The goal of this paper is to demonstrate and address challenges related to all aspects of performing a complete uncertainty quantification analysis of a complicated physics-based simulation like a 2D slab burner direct numerical simulation (DNS). The UQ framework includes the development of data-driven surrogate models, propagation of parametric uncertainties to the fuel regression rate--the primary quantity of interest--and Bayesian calibration of the latent heat of sublimation and a chemical reaction temperature exponent using experimental data. Two surrogate models, a Gaussian Process (GP) and a Hierarchical Multiscale Surrogate (HMS) were constructed using an ensemble of 64 simulations generated via Latin Hypercube sampling. HMS is superior for prediction demonstrated by cross-validation and able to achieve an error < 15% when predicting multiscale boundary quantities just from a few far field inputs. Subsequent Bayesian calibration of chemical kinetics and fuel response parameters against experimental observations showed that the default values used in the DNS should be higher to better match measurements. This study highlights the importance of surrogate model selection and parameter calibration in quantifying uncertainty in predictions of fuel regression rates in complex combustion systems.
- [180] arXiv:2411.19346 (replaced) [pdf, html, other]
-
Title: CLIP meets DINO for Tuning Zero-Shot Classifier using Unlabeled Image CollectionsMohamed Fazli Imam, Rufael Fedaku Marew, Jameel Hassan, Mustansar Fiaz, Alham Fikri Aji, Hisham CholakkalSubjects: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL); Machine Learning (cs.LG)
In the era of foundation models, CLIP has emerged as a powerful tool for aligning text & visual modalities into a common embedding space. However, the alignment objective used to train CLIP often results in subpar visual features for fine-grained tasks. In contrast, SSL-pretrained models like DINO excel at extracting rich visual features due to their specialized training paradigm. Yet, these SSL models require an additional supervised linear probing step, which relies on fully labeled data which is often expensive and difficult to obtain at scale. In this paper, we propose a label-free prompt-tuning method that leverages the rich visual features of self-supervised learning models (DINO) and the broad textual knowledge of large language models (LLMs) to largely enhance CLIP-based image classification performance using unlabeled images. Our approach unfolds in three key steps: (1) We generate robust textual feature embeddings that more accurately represent object classes by leveraging class-specific descriptions from LLMs, enabling more effective zero-shot classification compared to CLIP's default name-specific prompts. (2) These textual embeddings are then used to produce pseudo-labels to train an alignment module that integrates the complementary strengths of LLM description-based textual embeddings & DINO's visual features. (3) Finally, we prompt-tune CLIP's vision encoder through DINO-assisted supervision using the trained alignment module. This three-step process allows us to harness the best of visual & textual foundation models, resulting in a powerful and efficient approach that surpasses state-of-the-art label-free classification methods. Notably, our framework, NoLA (No Labels Attached), achieves an average absolute gain of 3.6% over the state-of-the-art LaFTer across 11 diverse image classification datasets. Our code & models can be found at this https URL.
- [181] arXiv:2411.19379 (replaced) [pdf, html, other]
-
Title: Marconi: Prefix Caching for the Era of Hybrid LLMsComments: MLSys 2025 camera-ready versionSubjects: Distributed, Parallel, and Cluster Computing (cs.DC); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Hybrid models that combine the language modeling capabilities of Attention layers with the efficiency of Recurrent layers (e.g., State Space Models) have gained traction in practically supporting long contexts in Large Language Model serving. Yet, the unique properties of these models complicate the usage of complementary efficiency optimizations such as prefix caching that skip redundant computations across requests. Most notably, their use of in-place state updates for recurrent layers precludes rolling back cache entries for partial sequence overlaps, and instead mandates only exact-match cache hits; the effect is a deluge of (large) cache entries per sequence, most of which yield minimal reuse opportunities. We present Marconi, the first system that supports efficient prefix caching with Hybrid LLMs. Key to Marconi are its novel admission and eviction policies that more judiciously assess potential cache entries based not only on recency, but also on (1) forecasts of their reuse likelihood across a taxonomy of different hit scenarios, and (2) the compute savings that hits deliver relative to memory footprints. Across diverse workloads and Hybrid models, Marconi achieves up to 34.4$\times$ higher token hit rates (71.1% or 617 ms lower TTFT) compared to state-of-the-art prefix caching systems.
- [182] arXiv:2412.11943 (replaced) [pdf, html, other]
-
Title: autrainer: A Modular and Extensible Deep Learning Toolkit for Computer Audition TasksSubjects: Sound (cs.SD); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Audio and Speech Processing (eess.AS)
This work introduces the key operating principles for autrainer, our new deep learning training framework for computer audition tasks. autrainer is a PyTorch-based toolkit that allows for rapid, reproducible, and easily extensible training on a variety of different computer audition tasks. Concretely, autrainer offers low-code training and supports a wide range of neural networks as well as preprocessing routines. In this work, we present an overview of its inner workings and key capabilities.
- [183] arXiv:2412.14719 (replaced) [pdf, html, other]
-
Title: Prototypical Calibrating Ambiguous Samples for Micro-Action RecognitionComments: Fix typos; Accepted by AAAI 2025Subjects: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
Micro-Action Recognition (MAR) has gained increasing attention due to its crucial role as a form of non-verbal communication in social interactions, with promising potential for applications in human communication and emotion analysis. However, current approaches often overlook the inherent ambiguity in micro-actions, which arises from the wide category range and subtle visual differences between categories. This oversight hampers the accuracy of micro-action recognition. In this paper, we propose a novel Prototypical Calibrating Ambiguous Network (PCAN) to unleash and mitigate the ambiguity of MAR. Firstly, we employ a hierarchical action-tree to identify the ambiguous sample, categorizing them into distinct sets of ambiguous samples of false negatives and false positives, considering both body- and action-level categories. Secondly, we implement an ambiguous contrastive refinement module to calibrate these ambiguous samples by regulating the distance between ambiguous samples and their corresponding prototypes. This calibration process aims to pull false negative (FN) samples closer to their respective prototypes and push false positive (FP) samples apart from their affiliated prototypes. In addition, we propose a new prototypical diversity amplification loss to strengthen the model's capacity by amplifying the differences between different prototypes. Finally, we propose a prototype-guided rectification to rectify prediction by incorporating the representability of prototypes. Extensive experiments conducted on the benchmark dataset demonstrate the superior performance of our method compared to existing approaches. The code is available at this https URL.
- [184] arXiv:2501.06425 (replaced) [pdf, html, other]
-
Title: Tensor Product Attention Is All You NeedComments: 31 pages, 6 figuresSubjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Scaling language models to handle longer input sequences typically necessitates large key-value (KV) caches, resulting in substantial memory overhead during inference. In this paper, we propose Tensor Product Attention (TPA), a novel attention mechanism that uses tensor decompositions to represent queries, keys, and values compactly, significantly shrinking KV cache size at inference time. By factorizing these representations into contextual low-rank components (contextual factorization) and seamlessly integrating with RoPE, TPA achieves improved model quality alongside memory efficiency. Based on TPA, we introduce the Tensor ProducT ATTenTion Transformer (T6), a new model architecture for sequence modeling. Through extensive empirical evaluation of language modeling tasks, we demonstrate that T6 exceeds the performance of standard Transformer baselines including MHA, MQA, GQA, and MLA across various metrics, including perplexity and a range of renowned evaluation benchmarks. Notably, TPA's memory efficiency enables the processing of significantly longer sequences under fixed resource constraints, addressing a critical scalability challenge in modern language models. The code is available at this https URL.
- [185] arXiv:2501.07824 (replaced) [pdf, html, other]
-
Title: Real-time Verification and Refinement of Language Model Text GenerationSubjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Large language models (LLMs) have shown remarkable performance across a wide range of natural language tasks. However, a critical challenge remains in that they sometimes generate factually incorrect answers. To address this, while many previous work has focused on identifying errors in their generation and further refining them, they are slow in deployment since they are designed to verify the response from LLMs only after their entire generation (from the first to last tokens) is done. Further, we observe that once LLMs generate incorrect tokens early on, there is a higher likelihood that subsequent tokens will also be factually incorrect. To this end, in this work, we propose Streaming-VR (Streaming Verification and Refinement), a novel approach designed to enhance the efficiency of verification and refinement of LLM outputs. Specifically, the proposed Streaming-VR enables on-the-fly verification and correction of tokens as they are being generated, similar to a streaming process, ensuring that each subset of tokens is checked and refined in real-time by another LLM as the LLM constructs its response. Through comprehensive evaluations on multiple datasets, we demonstrate that our approach not only enhances the factual accuracy of LLMs, but also offers a more efficient solution compared to prior refinement methods.
- [186] arXiv:2501.14174 (replaced) [pdf, html, other]
-
Title: Dreamweaver: Learning Compositional World Models from PixelsSubjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Humans have an innate ability to decompose their perceptions of the world into objects and their attributes, such as colors, shapes, and movement patterns. This cognitive process enables us to imagine novel futures by recombining familiar concepts. However, replicating this ability in artificial intelligence systems has proven challenging, particularly when it comes to modeling videos into compositional concepts and generating unseen, recomposed futures without relying on auxiliary data, such as text, masks, or bounding boxes. In this paper, we propose Dreamweaver, a neural architecture designed to discover hierarchical and compositional representations from raw videos and generate compositional future simulations. Our approach leverages a novel Recurrent Block-Slot Unit (RBSU) to decompose videos into their constituent objects and attributes. In addition, Dreamweaver uses a multi-future-frame prediction objective to capture disentangled representations for dynamic concepts more effectively as well as static concepts. In experiments, we demonstrate our model outperforms current state-of-the-art baselines for world modeling when evaluated under the DCI framework across multiple datasets. Furthermore, we show how the modularized concept representations of our model enable compositional imagination, allowing the generation of novel videos by recombining attributes from previously seen objects. this http URL
- [187] arXiv:2503.01284 (replaced) [pdf, html, other]
-
Title: Soybean Disease Detection via Interpretable Hybrid CNN-GNN: Integrating MobileNetV2 and GraphSAGE with Cross-Modal AttentionSubjects: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
Soybean leaf disease detection is critical for agricultural productivity but faces challenges due to visually similar symptoms and limited interpretability in conventional methods. While Convolutional Neural Networks (CNNs) excel in spatial feature extraction, they often neglect inter-image relational dependencies, leading to misclassifications. This paper proposes an interpretable hybrid Sequential CNN-Graph Neural Network (GNN) framework that synergizes MobileNetV2 for localized feature extraction and GraphSAGE for relational modeling. The framework constructs a graph where nodes represent leaf images, with edges defined by cosine similarity-based adjacency matrices and adaptive neighborhood sampling. This design captures fine-grained lesion features and global symptom patterns, addressing inter-class similarity challenges. Cross-modal interpretability is achieved via Grad-CAM and Eigen-CAM visualizations, generating heatmaps to highlight disease-influential regions. Evaluated on a dataset of ten soybean leaf diseases, the model achieves $97.16\%$ accuracy, surpassing standalone CNNs ($\le95.04\%$) and traditional machine learning models ($\le77.05\%$). Ablation studies validate the sequential architecture's superiority over parallel or single-model configurations. With only 2.3 million parameters, the lightweight MobileNetV2-GraphSAGE combination ensures computational efficiency, enabling real-time deployment in resource-constrained environments. The proposed approach bridges the gap between accurate classification and practical applicability, offering a robust, interpretable tool for agricultural diagnostics while advancing CNN-GNN integration in plant pathology research.
- [188] arXiv:2503.17845 (replaced) [pdf, html, other]
-
Title: Graphical Transformation ModelsComments: 36 pages, 10 Figures, presented at the DAGStat 2025 in BerlinSubjects: Methodology (stat.ME); Machine Learning (cs.LG); Machine Learning (stat.ML)
Graphical Transformation Models (GTMs) are introduced as a novel approach to effectively model multivariate data with intricate marginals and complex dependency structures non-parametrically, while maintaining interpretability through the identification of varying conditional independencies. GTMs extend multivariate transformation models by replacing the Gaussian copula with a custom-designed multivariate transformation, offering two major advantages. Firstly, GTMs can capture more complex interdependencies using penalized splines, which also provide an efficient regularization scheme. Secondly, we demonstrate how to approximately regularize GTMs using a lasso penalty towards pairwise conditional independencies, akin to Gaussian graphical models. The model's robustness and effectiveness are validated through simulations, showcasing its ability to accurately learn parametric vine copulas and identify conditional independencies. Additionally, the model is applied to a benchmark astrophysics dataset, where the GTM demonstrates favorable performance compared to non-parametric vine copulas in learning complex multivariate distributions.
- [189] arXiv:2503.19190 (replaced) [pdf, html, other]
-
Title: Universal Architectures for the Learning of Polyhedral Norms and Convex RegularizersSubjects: Machine Learning (stat.ML); Machine Learning (cs.LG); Optimization and Control (math.OC)
This paper addresses the task of learning convex regularizers to guide the reconstruction of images from limited data. By imposing that the reconstruction be amplitude-equivariant, we narrow down the class of admissible functionals to those that can be expressed as a power of a seminorm. We then show that such functionals can be approximated to arbitrary precision with the help of polyhedral norms. In particular, we identify two dual parameterizations of such systems: (i) a synthesis form with an $\ell_1$-penalty that involves some learnable dictionary; and (ii) an analysis form with an $\ell_\infty$-penalty that involves a trainable regularization operator. After having provided geometric insights and proved that the two forms are universal, we propose an implementation that relies on a specific architecture (tight frame with a weighted $\ell_1$ penalty) that is easy to train. We illustrate its use for denoising and the reconstruction of biomedical images. We find that the proposed framework outperforms the sparsity-based methods of compressed sensing, while it offers essentially the same convergence and robustness guarantees.
- [190] arXiv:2503.19949 (replaced) [pdf, html, other]
-
Title: Automated Video-EEG Analysis in Epilepsy Studies: Advances and ChallengesSubjects: Image and Video Processing (eess.IV); Machine Learning (cs.LG)
Epilepsy is typically diagnosed through electroencephalography (EEG) and long-term video-EEG (vEEG) monitoring. The manual analysis of vEEG recordings is time-consuming, necessitating automated tools for seizure detection. Recent advancements in machine learning have shown promise in real-time seizure detection and prediction using EEG and video data. However, diversity of seizure symptoms, markup ambiguities, and limited availability of multimodal datasets hinder progress. This paper reviews the latest developments in automated video-EEG analysis and discusses the integration of multimodal data. We also propose a novel pipeline for treatment effect estimation from vEEG data using concept-based learning, offering a pathway for future research in this domain.
- [191] arXiv:2504.02670 (replaced) [pdf, html, other]
-
Title: Affordable AI Assistants with Knowledge Graph of ThoughtsMaciej Besta, Lorenzo Paleari, Jia Hao Andrea Jiang, Robert Gerstenberger, You Wu, Patrick Iff, Ales Kubicek, Piotr Nyczyk, Diana Khimey, Jón Gunnar Hannesson, Grzegorz Kwaśniewski, Marcin Copik, Hubert Niewiadomski, Torsten HoeflerSubjects: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Information Retrieval (cs.IR); Machine Learning (cs.LG)
Large Language Models (LLMs) are revolutionizing the development of AI assistants capable of performing diverse tasks across domains. However, current state-of-the-art LLM-driven agents face significant challenges, including high operational costs and limited success rates on complex benchmarks like GAIA. To address these issues, we propose the Knowledge Graph of Thoughts (KGoT), an innovative AI assistant architecture that integrates LLM reasoning with dynamically constructed knowledge graphs (KGs). KGoT extracts and structures task-relevant knowledge into a dynamic KG representation, iteratively enhanced through external tools such as math solvers, web crawlers, and Python scripts. Such structured representation of task-relevant knowledge enables low-cost models to solve complex tasks effectively. For example, KGoT achieves a 29% improvement in task success rates on the GAIA benchmark compared to Hugging Face Agents with GPT-4o mini, while reducing costs by over 36x compared to GPT-4o. Improvements for recent reasoning models are similar, e.g., 36% and 37.5% for Qwen2.5-32B and Deepseek-R1-70B, respectively. KGoT offers a scalable, affordable, and high-performing solution for AI assistants.
- [192] arXiv:2504.03624 (replaced) [pdf, html, other]
-
Title: Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer ModelsNVIDIA: Aaron Blakeman, Aarti Basant, Abhinav Khattar, Adithya Renduchintala, Akhiad Bercovich, Aleksander Ficek, Alexis Bjorlin, Ali Taghibakhshi, Amala Sanjay Deshmukh, Ameya Sunil Mahabaleshwarkar, Andrew Tao, Anna Shors, Ashwath Aithal, Ashwin Poojary, Ayush Dattagupta, Balaram Buddharaju, Bobby Chen, Boris Ginsburg, Boxin Wang, Brandon Norick, Brian Butterfield, Bryan Catanzaro, Carlo del Mundo, Chengyu Dong, Christine Harvey, Christopher Parisien, Dan Su, Daniel Korzekwa, Danny Yin, Daria Gitman, David Mosallanezhad, Deepak Narayanan, Denys Fridman, Dima Rekesh, Ding Ma, Dmytro Pykhtar, Dong Ahn, Duncan Riach, Dusan Stosic, Eileen Long, Elad Segal, Ellie Evans, Eric Chung, Erick Galinkin, Evelina Bakhturina, Ewa Dobrowolska, Fei Jia, Fuxiao Liu, Gargi Prasad, Gerald Shen, Guilin Liu, Guo Chen, Haifeng Qian, Helen Ngo, Hongbin Liu, Hui Li, Igor Gitman, Ilia Karmanov, Ivan Moshkov, Izik Golan, Jan Kautz, Jane Polak Scowcroft, Jared Casper, Jarno Seppanen, Jason Lu, Jason Sewall, Jiaqi Zeng, Jiaxuan You, Jimmy Zhang, Jing Zhang, Jining Huang, Jinze Xue, Jocelyn Huang, Joey Conway, John Kamalu, Jon Barker, Jonathan Cohen, Joseph Jennings, Jupinder Parmar, Karan Sapra, Kari Briski, Kateryna Chumachenko, Katherine Luna, Keshav Santhanam, Kezhi Kong, Kirthi Sivamani, Krzysztof Pawelec, Kumar Anik, Kunlun Li, Lawrence McAfee, Leon Derczynski, Lindsey Pavao, Luis Vega, Lukas Voegtle, Maciej Bala, Maer Rodrigues de Melo, Makesh Narsimhan Sreedhar, Marcin Chochowski, Markus KlieglSubjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
As inference-time scaling becomes critical for enhanced reasoning capabilities, it is increasingly becoming important to build models that are efficient to infer. We introduce Nemotron-H, a family of 8B and 56B/47B hybrid Mamba-Transformer models designed to reduce inference cost for a given accuracy level. To achieve this goal, we replace the majority of self-attention layers in the common Transformer model architecture with Mamba layers that perform constant computation and require constant memory per generated token. We show that Nemotron-H models offer either better or on-par accuracy compared to other similarly-sized state-of-the-art open-sourced Transformer models (e.g., Qwen-2.5-7B/72B and Llama-3.1-8B/70B), while being up to 3$\times$ faster at inference. To further increase inference speed and reduce the memory required at inference time, we created Nemotron-H-47B-Base from the 56B model using a new compression via pruning and distillation technique called MiniPuzzle. Nemotron-H-47B-Base achieves similar accuracy to the 56B model, but is 20% faster to infer. In addition, we introduce an FP8-based training recipe and show that it can achieve on par results with BF16-based training. This recipe is used to train the 56B model. All Nemotron-H models will be released, with support in Hugging Face, NeMo, and Megatron-LM.
- [193] arXiv:2504.03699 (replaced) [pdf, html, other]
-
Title: Reinforcing Clinical Decision Support through Multi-Agent Systems and Ethical AI GovernanceSubjects: Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Machine Learning (cs.LG); Multiagent Systems (cs.MA); Quantitative Methods (q-bio.QM)
In the age of data-driven medicine, it is paramount to include explainable and ethically managed artificial intelligence in explaining clinical decision support systems to achieve trustworthy and effective patient care. The focus of this paper is on a new architecture of a multi-agent system for clinical decision support that uses modular agents to analyze laboratory results, vital signs, and the clinical context and then integrates these results to drive predictions and validate outcomes. We describe our implementation with the eICU database to run lab-analysis-specific agents, vitals-only interpreters, and contextual reasoners and then run the prediction module and a validation agent. Everything is a transparent implementation of business logic, influenced by the principles of ethical AI governance such as Autonomy, Fairness, and Accountability. It provides visible results that this agent-based framework not only improves on interpretability and accuracy but also on reinforcing trust in AI-assisted decisions in an intensive care setting.
- [194] arXiv:2504.04372 (replaced) [pdf, html, other]
-
Title: How Accurately Do Large Language Models Understand Code?Sabaat Haroon, Ahmad Faraz Khan, Ahmad Humayun, Waris Gill, Abdul Haddi Amjad, Ali R. Butt, Mohammad Taha Khan, Muhammad Ali GulzarComments: This paper is currently Under Review. It consists of 11 pages, 12 Figures, and 5 TablesSubjects: Software Engineering (cs.SE); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Large Language Models (LLMs) are increasingly used in post-development tasks such as code repair and testing. A key factor in these tasks' success is the model's deep understanding of code. However, the extent to which LLMs truly understand code remains largely unevaluated. Quantifying code comprehension is challenging due to its abstract nature and the lack of a standardized metric. Previously, this was assessed through developer surveys, which are not feasible for evaluating LLMs. Existing LLM benchmarks focus primarily on code generation, fundamentally different from code comprehension. Additionally, fixed benchmarks quickly become obsolete as they become part of the training data. This paper presents the first large-scale empirical investigation into LLMs' ability to understand code. Inspired by mutation testing, we use an LLM's fault-finding ability as a proxy for its deep code understanding. This approach is based on the insight that a model capable of identifying subtle functional discrepancies must understand the code well. We inject faults in real-world programs and ask the LLM to localize them, ensuring the specifications suffice for fault localization. Next, we apply semantic-preserving code mutations (SPMs) to the faulty programs and test whether the LLMs still locate the faults, verifying their confidence in code understanding. We evaluate nine popular LLMs on 600,010 debugging tasks from 670 Java and 637 Python programs. We find that LLMs lose the ability to debug the same bug in 78% of faulty programs when SPMs are applied, indicating a shallow understanding of code and reliance on features irrelevant to semantics. We also find that LLMs understand code earlier in the program better than later. This suggests that LLMs' code comprehension remains tied to lexical and syntactic features due to tokenization designed for natural languages, which overlooks code semantics.
- [195] arXiv:2504.05500 (replaced) [pdf, html, other]
-
Title: Prism: Dynamic and Flexible Benchmarking of LLMs Code Generation with Monte Carlo Tree SearchSubjects: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Software Engineering (cs.SE)
The rapid advancement of Large Language Models (LLMs) has outpaced traditional evaluation methods. Static benchmarks fail to capture the depth and breadth of LLM capabilities and eventually become obsolete, while most dynamic approaches either rely too heavily on LLM-based evaluation or remain constrained by predefined test sets. We introduce Prism, a flexible, dynamic benchmarking framework designed for comprehensive LLM assessment. Prism builds on three key components: (1) a tree-based state representation that models evaluation as a Markov Decision Process, (2) a Monte Carlo Tree Search algorithm adapted to uncover challenging evaluation scenarios, and (3) a multi-agent evaluation pipeline that enables simultaneous assessment of diverse capabilities. To ensure robust evaluation, Prism integrates structural measurements of tree exploration patterns with performance metrics across difficulty levels, providing detailed diagnostics of error patterns, test coverage, and solution approaches. Through extensive experiments on five state-of-the-art LLMs, we analyze how model architecture and scale influence code generation performance across varying task difficulties. Our results demonstrate Prism's effectiveness as a dynamic benchmark that evolves with model advancements while offering deeper insights into their limitations.
- [196] arXiv:2504.06293 (replaced) [pdf, html, other]
-
Title: Generative AI Enhanced Financial Risk Management Information RetrievalComments: 10 pages, 3 figures, 2 tables, 1 equationSubjects: Risk Management (q-fin.RM); Machine Learning (cs.LG)
Risk management in finance involves recognizing, evaluating, and addressing financial risks to maintain stability and ensure regulatory compliance. Extracting relevant insights from extensive regulatory documents is a complex challenge requiring advanced retrieval and language models. This paper introduces RiskData, a dataset specifically curated for finetuning embedding models in risk management, and RiskEmbed, a finetuned embedding model designed to improve retrieval accuracy in financial question-answering systems. The dataset is derived from 94 regulatory guidelines published by the Office of the Superintendent of Financial Institutions (OSFI) from 1991 to 2024. We finetune a state-of-the-art sentence BERT embedding model to enhance domain-specific retrieval performance typically for Retrieval-Augmented Generation (RAG) systems. Experimental results demonstrate that RiskEmbed significantly outperforms general-purpose and financial embedding models, achieving substantial improvements in ranking metrics. By open-sourcing both the dataset and the model, we provide a valuable resource for financial institutions and researchers aiming to develop more accurate and efficient risk management AI solutions.
