Formatting Instructions for ICLR 2025
Conference Submissions

Antiquus S. Hippocampus, Natalia Cerebro & Amelie P. Amygdale
Department of Computer Science
Cranberry-Lemon University
Pittsburgh, PA 15213, USA
{hippo,brain,jen}@cs.cranberry-lemon.edu
&Ji Q. Ren & Yevgeny LeNet
Department of Computational Neuroscience
University of the Witwatersrand
Joburg, South Africa
{robot,net}@wits.ac.za
\ANDCoauthor
Affiliation
Address
email
Use footnote for providing further information about author (webpage, alternative address)—not for acknowledging funding agencies. Funding acknowledgements go at the end of the paper.
Abstract

The abstract paragraph should be indented 1/2 inch (3 picas) on both left and right-hand margins. Use 10 point type, with a vertical spacing of 11 points. The word Abstract must be centered, in small caps, and in point size 12. Two line spaces precede the abstract. The abstract must be limited to one paragraph.

1 Submission of conference papers to ICLR 2025

ICLR requires electronic submissions, processed by https://openreview.net/. See ICLR’s website for more instructions.

If your paper is ultimately accepted, the statement \iclrfinalcopy should be inserted to adjust the format to the camera ready requirements.

The format for the submissions is a variant of the NeurIPS format. Please read carefully the instructions below, and follow them faithfully.

1.1 Style

Papers to be submitted to ICLR 2025 must be prepared according to the instructions presented here.

Authors are required to use the ICLR style files obtainable at the ICLR website. Please make sure you use the current files and not previous versions. Tweaking the style files may be grounds for rejection.

1.2 Retrieval of style files

The style files for ICLR and other conference information are available online at:

http://www.iclr.cc/

The file iclr2025_conference.pdf contains these instructions and illustrates the various formatting requirements your ICLR paper must satisfy. Submissions must be made using and the style files iclr2025_conference.sty and iclr2025_conference.bst (to be used with 2e). The file iclr2025_conference.tex may be used as a “shell” for writing your paper. All you have to do is replace the author, title, abstract, and text of the paper with your own.

The formatting instructions contained in these style files are summarized in sections 2, 3, and 4 below.

2 General formatting instructions

The text must be confined within a rectangle 5.5 inches (33 picas) wide and 9 inches (54 picas) long. The left margin is 1.5 inch (9 picas). Use 10 point type with a vertical spacing of 11 points. Times New Roman is the preferred typeface throughout. Paragraphs are separated by 1/2 line space, with no indentation.

Paper title is 17 point, in small caps and left-aligned. All pages should start at 1 inch (6 picas) from the top of the page.

Authors’ names are set in boldface, and each name is placed above its corresponding address. The lead author’s name is to be listed first, and the co-authors’ names are set to follow. Authors sharing the same address can be on the same line.

Please pay special attention to the instructions in section 4 regarding figures, tables, acknowledgments, and references.

There will be a strict upper limit of 10 pages for the main text of the initial submission, with unlimited additional pages for citations.

3 Headings: first level

First level headings are in small caps, flush left and in point size 12. One line space before the first level heading and 1/2 line space after the first level heading.

3.1 Headings: second level

Second level headings are in small caps, flush left and in point size 10. One line space before the second level heading and 1/2 line space after the second level heading.

3.1.1 Headings: third level

Third level headings are in small caps, flush left and in point size 10. One line space before the third level heading and 1/2 line space after the third level heading.

4 Citations, figures, tables, references

These instructions apply to everyone, regardless of the formatter being used.

4.1 Citations within the text

Citations within the text should be based on the natbib package and include the authors’ last names and year (with the “et al.” construct for more than two authors). When the authors or the publication are included in the sentence, the citation should not be in parenthesis using \citet{} (as in “See Hinton06 for more information.”). Otherwise, the citation should be in parenthesis using \citep{} (as in “Deep learning shows promise to make progress towards AI (Bengio+chapter2007).”).

The corresponding references are to be listed in alphabetical order of authors, in the References section. As to the format of the references themselves, any style is acceptable as long as it is used consistently.

4.2 Footnotes

Indicate footnotes with a number111Sample of the first footnote in the text. Place the footnotes at the bottom of the page on which they appear. Precede the footnote with a horizontal rule of 2 inches (12 picas).222Sample of the second footnote

4.3 Figures

All artwork must be neat, clean, and legible. Lines should be dark enough for purposes of reproduction; art work should not be hand-drawn. The figure number and caption always appear after the figure. Place one line space before the figure caption, and one line space after the figure. The figure caption is lower case (except for first word and proper nouns); figures are numbered consecutively.

Make sure the figure caption does not get separated from the figure. Leave sufficient space to avoid splitting the figure and figure caption.

You may use color figures. However, it is best for the figure captions and the paper body to make sense if the paper is printed either in black/white or in color.

 


Figure 1: Sample figure caption.

4.4 Tables

All tables must be centered, neat, clean and legible. Do not use hand-drawn tables. The table number and title always appear before the table. See Table 1.

Place one line space before the table title, one line space after the table title, and one line space after the table. The table title must be lower case (except for first word and proper nouns); tables are numbered consecutively.

Table 1: Sample table title
PART DESCRIPTION
Dendrite Input terminal
Axon Output terminal
Soma Cell body (contains cell nucleus)

5 Default Notation

In an attempt to encourage standardized notation, we have included the notation file from the textbook, Deep Learning goodfellow2016deep available at https://github.com/goodfeli/dlbook_notation/. Use of this style is not required and can be disabled by commenting out math_commands.tex.

Numbers and Arrays

a𝑎\displaystyle aitalic_a A scalar (integer or real)
𝒂𝒂\displaystyle{\bm{a}}bold_italic_a A vector
𝑨𝑨\displaystyle{\bm{A}}bold_italic_A A matrix
𝑨𝑨\displaystyle{\bm{\mathsfit{A}}}bold_slanted_A A tensor
𝑰nsubscript𝑰𝑛\displaystyle{\bm{I}}_{n}bold_italic_I start_POSTSUBSCRIPT italic_n end_POSTSUBSCRIPT Identity matrix with n𝑛nitalic_n rows and n𝑛nitalic_n columns
𝑰𝑰\displaystyle{\bm{I}}bold_italic_I Identity matrix with dimensionality implied by context
𝒆(i)superscript𝒆𝑖\displaystyle{\bm{e}}^{(i)}bold_italic_e start_POSTSUPERSCRIPT ( italic_i ) end_POSTSUPERSCRIPT Standard basis vector [0,,0,1,0,,0]00100[0,\dots,0,1,0,\dots,0][ 0 , … , 0 , 1 , 0 , … , 0 ] with a 1 at position i𝑖iitalic_i
diag(𝒂)diag𝒂\displaystyle\text{diag}({\bm{a}})diag ( bold_italic_a ) A square, diagonal matrix with diagonal entries given by 𝒂𝒂{\bm{a}}bold_italic_a
a A scalar random variable
𝐚𝐚\displaystyle{\mathbf{a}}bold_a A vector-valued random variable
𝐀𝐀\displaystyle{\mathbf{A}}bold_A A matrix-valued random variable

Sets and Graphs

𝔸𝔸\displaystyle{\mathbb{A}}blackboard_A A set
\displaystyle\mathbb{R}blackboard_R The set of real numbers
{0,1}01\displaystyle\{0,1\}{ 0 , 1 } The set containing 0 and 1
{0,1,,n}01𝑛\displaystyle\{0,1,\dots,n\}{ 0 , 1 , … , italic_n } The set of all integers between 00 and n𝑛nitalic_n
[a,b]𝑎𝑏\displaystyle[a,b][ italic_a , italic_b ] The real interval including a𝑎aitalic_a and b𝑏bitalic_b
(a,b]𝑎𝑏\displaystyle(a,b]( italic_a , italic_b ] The real interval excluding a𝑎aitalic_a but including b𝑏bitalic_b
𝔸\𝔹\𝔸𝔹\displaystyle{\mathbb{A}}\backslash{\mathbb{B}}blackboard_A \ blackboard_B Set subtraction, i.e., the set containing the elements of 𝔸𝔸{\mathbb{A}}blackboard_A that are not in 𝔹𝔹{\mathbb{B}}blackboard_B
𝒢𝒢\displaystyle{\mathcal{G}}caligraphic_G A graph
Pa𝒢(xi)𝑃subscript𝑎𝒢subscriptx𝑖\displaystyle Pa_{\mathcal{G}}({\textnormal{x}}_{i})italic_P italic_a start_POSTSUBSCRIPT caligraphic_G end_POSTSUBSCRIPT ( x start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ) The parents of xisubscriptx𝑖{\textnormal{x}}_{i}x start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT in 𝒢𝒢{\mathcal{G}}caligraphic_G

Indexing

aisubscript𝑎𝑖\displaystyle{a}_{i}italic_a start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT Element i𝑖iitalic_i of vector 𝒂𝒂{\bm{a}}bold_italic_a, with indexing starting at 1
aisubscript𝑎𝑖\displaystyle{a}_{-i}italic_a start_POSTSUBSCRIPT - italic_i end_POSTSUBSCRIPT All elements of vector 𝒂𝒂{\bm{a}}bold_italic_a except for element i𝑖iitalic_i
Ai,jsubscript𝐴𝑖𝑗\displaystyle{A}_{i,j}italic_A start_POSTSUBSCRIPT italic_i , italic_j end_POSTSUBSCRIPT Element i,j𝑖𝑗i,jitalic_i , italic_j of matrix 𝑨𝑨{\bm{A}}bold_italic_A
𝑨i,:subscript𝑨𝑖:\displaystyle{\bm{A}}_{i,:}bold_italic_A start_POSTSUBSCRIPT italic_i , : end_POSTSUBSCRIPT Row i𝑖iitalic_i of matrix 𝑨𝑨{\bm{A}}bold_italic_A
𝑨:,isubscript𝑨:𝑖\displaystyle{\bm{A}}_{:,i}bold_italic_A start_POSTSUBSCRIPT : , italic_i end_POSTSUBSCRIPT Column i𝑖iitalic_i of matrix 𝑨𝑨{\bm{A}}bold_italic_A
Ai,j,ksubscript𝐴𝑖𝑗𝑘\displaystyle{\mathsfit{A}}_{i,j,k}slanted_A start_POSTSUBSCRIPT italic_i , italic_j , italic_k end_POSTSUBSCRIPT Element (i,j,k)𝑖𝑗𝑘(i,j,k)( italic_i , italic_j , italic_k ) of a 3-D tensor 𝑨𝑨{\bm{\mathsfit{A}}}bold_slanted_A
𝑨:,:,isubscript𝑨::𝑖\displaystyle{\bm{\mathsfit{A}}}_{:,:,i}bold_slanted_A start_POSTSUBSCRIPT : , : , italic_i end_POSTSUBSCRIPT 2-D slice of a 3-D tensor
aisubscripta𝑖\displaystyle{\textnormal{a}}_{i}a start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT Element i𝑖iitalic_i of the random vector 𝐚𝐚{\mathbf{a}}bold_a

Calculus

dydx𝑑𝑦𝑑𝑥\displaystyle\frac{dy}{dx}divide start_ARG italic_d italic_y end_ARG start_ARG italic_d italic_x end_ARG Derivative of y𝑦yitalic_y with respect to x𝑥xitalic_x
yx𝑦𝑥\displaystyle\frac{\partial y}{\partial x}divide start_ARG ∂ italic_y end_ARG start_ARG ∂ italic_x end_ARG Partial derivative of y𝑦yitalic_y with respect to x𝑥xitalic_x
𝒙ysubscript𝒙𝑦\displaystyle\nabla_{\bm{x}}y∇ start_POSTSUBSCRIPT bold_italic_x end_POSTSUBSCRIPT italic_y Gradient of y𝑦yitalic_y with respect to 𝒙𝒙{\bm{x}}bold_italic_x
𝑿ysubscript𝑿𝑦\displaystyle\nabla_{\bm{X}}y∇ start_POSTSUBSCRIPT bold_italic_X end_POSTSUBSCRIPT italic_y Matrix derivatives of y𝑦yitalic_y with respect to 𝑿𝑿{\bm{X}}bold_italic_X
𝑿ysubscript𝑿𝑦\displaystyle\nabla_{\bm{\mathsfit{X}}}y∇ start_POSTSUBSCRIPT bold_slanted_X end_POSTSUBSCRIPT italic_y Tensor containing derivatives of y𝑦yitalic_y with respect to 𝑿𝑿{\bm{\mathsfit{X}}}bold_slanted_X
f𝒙𝑓𝒙\displaystyle\frac{\partial f}{\partial{\bm{x}}}divide start_ARG ∂ italic_f end_ARG start_ARG ∂ bold_italic_x end_ARG Jacobian matrix 𝑱m×n𝑱superscript𝑚𝑛{\bm{J}}\in\mathbb{R}^{m\times n}bold_italic_J ∈ blackboard_R start_POSTSUPERSCRIPT italic_m × italic_n end_POSTSUPERSCRIPT of f:nm:𝑓superscript𝑛superscript𝑚f:\mathbb{R}^{n}\rightarrow\mathbb{R}^{m}italic_f : blackboard_R start_POSTSUPERSCRIPT italic_n end_POSTSUPERSCRIPT → blackboard_R start_POSTSUPERSCRIPT italic_m end_POSTSUPERSCRIPT
𝒙2f(𝒙) or 𝑯(f)(𝒙)superscriptsubscript𝒙2𝑓𝒙 or 𝑯𝑓𝒙\displaystyle\nabla_{\bm{x}}^{2}f({\bm{x}})\text{ or }{\bm{H}}(f)({\bm{x}})∇ start_POSTSUBSCRIPT bold_italic_x end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_f ( bold_italic_x ) or bold_italic_H ( italic_f ) ( bold_italic_x ) The Hessian matrix of f𝑓fitalic_f at input point 𝒙𝒙{\bm{x}}bold_italic_x
f(𝒙)𝑑𝒙𝑓𝒙differential-d𝒙\displaystyle\int f({\bm{x}})d{\bm{x}}∫ italic_f ( bold_italic_x ) italic_d bold_italic_x Definite integral over the entire domain of 𝒙𝒙{\bm{x}}bold_italic_x
𝕊f(𝒙)𝑑𝒙subscript𝕊𝑓𝒙differential-d𝒙\displaystyle\int_{\mathbb{S}}f({\bm{x}})d{\bm{x}}∫ start_POSTSUBSCRIPT blackboard_S end_POSTSUBSCRIPT italic_f ( bold_italic_x ) italic_d bold_italic_x Definite integral with respect to 𝒙𝒙{\bm{x}}bold_italic_x over the set 𝕊𝕊{\mathbb{S}}blackboard_S

Probability and Information Theory

P(a)𝑃a\displaystyle P({\textnormal{a}})italic_P ( a ) A probability distribution over a discrete variable
p(a)𝑝a\displaystyle p({\textnormal{a}})italic_p ( a ) A probability distribution over a continuous variable, or over a variable whose type has not been specified
aPsimilar-toa𝑃\displaystyle{\textnormal{a}}\sim Pa ∼ italic_P Random variable a has distribution P𝑃Pitalic_P
𝔼xP[f(x)] or 𝔼f(x)subscript𝔼similar-tox𝑃delimited-[]𝑓𝑥 or 𝔼𝑓𝑥\displaystyle\mathbb{E}_{{\textnormal{x}}\sim P}[f(x)]\text{ or }\mathbb{E}f(x)blackboard_E start_POSTSUBSCRIPT x ∼ italic_P end_POSTSUBSCRIPT [ italic_f ( italic_x ) ] or blackboard_E italic_f ( italic_x ) Expectation of f(x)𝑓𝑥f(x)italic_f ( italic_x ) with respect to P(x)𝑃xP({\textnormal{x}})italic_P ( x )
Var(f(x))Var𝑓𝑥\displaystyle\mathrm{Var}(f(x))roman_Var ( italic_f ( italic_x ) ) Variance of f(x)𝑓𝑥f(x)italic_f ( italic_x ) under P(x)𝑃xP({\textnormal{x}})italic_P ( x )
Cov(f(x),g(x))Cov𝑓𝑥𝑔𝑥\displaystyle\mathrm{Cov}(f(x),g(x))roman_Cov ( italic_f ( italic_x ) , italic_g ( italic_x ) ) Covariance of f(x)𝑓𝑥f(x)italic_f ( italic_x ) and g(x)𝑔𝑥g(x)italic_g ( italic_x ) under P(x)𝑃xP({\textnormal{x}})italic_P ( x )
H(x)𝐻x\displaystyle H({\textnormal{x}})italic_H ( x ) Shannon entropy of the random variable x
DKL(PQ)subscript𝐷KLconditional𝑃𝑄\displaystyle D_{\mathrm{KL}}(P\|Q)italic_D start_POSTSUBSCRIPT roman_KL end_POSTSUBSCRIPT ( italic_P ∥ italic_Q ) Kullback-Leibler divergence of P and Q
𝒩(𝒙;𝝁,𝚺)𝒩𝒙𝝁𝚺\displaystyle\mathcal{N}({\bm{x}};{\bm{\mu}},{\bm{\Sigma}})caligraphic_N ( bold_italic_x ; bold_italic_μ , bold_Σ ) Gaussian distribution over 𝒙𝒙{\bm{x}}bold_italic_x with mean 𝝁𝝁{\bm{\mu}}bold_italic_μ and covariance 𝚺𝚺{\bm{\Sigma}}bold_Σ

Functions

f:𝔸𝔹:𝑓𝔸𝔹\displaystyle f:{\mathbb{A}}\rightarrow{\mathbb{B}}italic_f : blackboard_A → blackboard_B The function f𝑓fitalic_f with domain 𝔸𝔸{\mathbb{A}}blackboard_A and range 𝔹𝔹{\mathbb{B}}blackboard_B
fg𝑓𝑔\displaystyle f\circ gitalic_f ∘ italic_g Composition of the functions f𝑓fitalic_f and g𝑔gitalic_g
f(𝒙;𝜽)𝑓𝒙𝜽\displaystyle f({\bm{x}};{\bm{\theta}})italic_f ( bold_italic_x ; bold_italic_θ ) A function of 𝒙𝒙{\bm{x}}bold_italic_x parametrized by 𝜽𝜽{\bm{\theta}}bold_italic_θ. (Sometimes we write f(𝒙)𝑓𝒙f({\bm{x}})italic_f ( bold_italic_x ) and omit the argument 𝜽𝜽{\bm{\theta}}bold_italic_θ to lighten notation)
logx𝑥\displaystyle\log xroman_log italic_x Natural logarithm of x𝑥xitalic_x
σ(x)𝜎𝑥\displaystyle\sigma(x)italic_σ ( italic_x ) Logistic sigmoid, 11+exp(x)11𝑥\displaystyle\frac{1}{1+\exp(-x)}divide start_ARG 1 end_ARG start_ARG 1 + roman_exp ( - italic_x ) end_ARG
ζ(x)𝜁𝑥\displaystyle\zeta(x)italic_ζ ( italic_x ) Softplus, log(1+exp(x))1𝑥\log(1+\exp(x))roman_log ( 1 + roman_exp ( italic_x ) )
𝒙psubscriptnorm𝒙𝑝\displaystyle||{\bm{x}}||_{p}| | bold_italic_x | | start_POSTSUBSCRIPT italic_p end_POSTSUBSCRIPT Lpsuperscript𝐿𝑝L^{p}italic_L start_POSTSUPERSCRIPT italic_p end_POSTSUPERSCRIPT norm of 𝒙𝒙{\bm{x}}bold_italic_x
𝒙norm𝒙\displaystyle||{\bm{x}}||| | bold_italic_x | | L2superscript𝐿2L^{2}italic_L start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT norm of 𝒙𝒙{\bm{x}}bold_italic_x
x+superscript𝑥\displaystyle x^{+}italic_x start_POSTSUPERSCRIPT + end_POSTSUPERSCRIPT Positive part of x𝑥xitalic_x, i.e., max(0,x)0𝑥\max(0,x)roman_max ( 0 , italic_x )
𝟏conditionsubscript1condition\displaystyle\bm{1}_{\mathrm{condition}}bold_1 start_POSTSUBSCRIPT roman_condition end_POSTSUBSCRIPT is 1 if the condition is true, 0 otherwise

6 Final instructions

Do not change any aspects of the formatting parameters in the style files. In particular, do not modify the width or length of the rectangle the text should fit into, and do not change font sizes (except perhaps in the References section; see below). Please note that pages should be numbered.

7 Preparing PostScript or PDF files

Please prepare PostScript or PDF files with paper size “US Letter”, and not, for example, “A4”. The -t letter option on dvips will produce US Letter files.

Consider directly generating PDF files using pdflatex (especially if you are a MiKTeX user). PDF figures must be substituted for EPS figures, however.

Otherwise, please generate your PostScript and PDF files with the following commands:

dvips mypaper.dvi -t letter -Ppdf -G0 -o mypaper.ps
ps2pdf mypaper.ps mypaper.pdf

7.1 Margins in LaTeX

Most of the margin problems come from figures positioned by hand using \special or other commands. We suggest using the command \includegraphics from the graphicx package. Always specify the figure width as a multiple of the line width as in the example below using .eps graphics

   \usepackage[dvips]{graphicx} ...
   \includegraphics[width=0.8\linewidth]{myfile.eps}

or

   \usepackage[pdftex]{graphicx} ...
   \includegraphics[width=0.8\linewidth]{myfile.pdf}

for .pdf graphics. See section 4.4 in the graphics bundle documentation (http://www.ctan.org/tex-archive/macros/latex/required/graphics/grfguide.ps)

A number of width problems arise when LaTeX cannot properly hyphenate a line. Please give LaTeX hyphenation hints using the \- command.

Author Contributions

If you’d like to, you may include a section for author contributions as is done in many journals. This is optional and at the discretion of the authors.

Acknowledgments

Use unnumbered third level headings for the acknowledgments. All acknowledgments, including those to funding agencies, go at the end of the paper.

Appendix A Appendix

You may include other additional sections here.