iclr-paper

name

iclr-paper

description

Use this skill to help write, structure, polish, format, or review a paper for ICLR (International Conference on Learning Representations, any year), or whenever the user mentions 'ICLR', an OpenReview deep-learning submission, the iclr single-column style, the reproducibility statement, or the ICLR rebuttal/open-review process. Helps with ideation and narrative, section structure, sentence-level polishing, ICLR LaTeX formatting, figures and tables (including leaving sized placeholders for figures), citation verification, rebuttal preparation, and a submission checklist. Double-blind, public-review venue. Verify exact dates, page limits, and policies against that year's official ICLR Call for Papers.

ICLR Paper Writing

Help the user take a deep-learning paper from idea to a polished, correctly formatted, double-blind ICLR submission, and through the public rebuttal phase. You can brainstorm the narrative, structure sections, polish prose, build LaTeX/figures/tables (or leave sized placeholders), verify citations, and run a reviewer self-check.

How to use this skill

Load the matching reference file on demand (not all at once):

User wants	Do this	Reference
Find the story / contribution / outline	nail the one-sentence contribution, then outline	references/ideation-and-structure.md
Draft or restructure a section	use the section blueprint	references/ideation-and-structure.md
Polish / tighten / "does this flow"	apply clarity principles + claim-evidence map	references/writing-and-polishing.md
Figures, tables, or placeholders	design or reserve space; generate plots	references/figures-and-tables.md, assets/
Generate a flowchart / pipeline / experiment plot	run the helper scripts	references/generating-diagrams.md, assets/make_flowchart.py, assets/make_experiment_figures.py
Add citations	verify before citing; never hallucinate	references/citations.md
Set up LaTeX / fix formatting / convert venue	template workflow + pitfalls	references/latex-and-submission.md
Final pass before submitting	run the checklist below + 5-dimension self-review	this file + references/writing-and-polishing.md

Be proactive: if the contribution and results are clear, deliver a full draft and flag open choices, rather than asking permission section by section.

At a glance

• Conference: International Conference on Learning Representations (annual).
• Submission system: OpenReview — review is double-blind but reviews and rebuttals are publicly visible, with an author-reviewer discussion period.
• Write to anticipate reviewer questions and leave room to answer them in rebuttal.

Dates change every year — check the current CFP

Do not rely on memorized dates. For the target cycle, open https://iclr.cc and read off the abstract deadline, full-paper deadline, the rebuttal/discussion window, notification, camera-ready, and conference dates/location. Rough rhythm (planning only): abstract and full paper around late September of the prior year; conference the following spring.

Format (ICLR-specific)

• Template: official ICLR style for the target year (e.g. iclrYYYY_conference.sty), single column, 10pt.
• Length: about 9 pages of main text (recent cycles), with unlimited references and appendix; camera-ready usually gets one extra page. Confirm the exact limit in the CFP.
• A reproducibility statement is expected; an (anonymized) code link helps scores.
• Recent cycles have required an LLM-usage disclosure; check the current CFP.

Structure notes specific to ICLR

Follow the general blueprint in references/ideation-and-structure.md, with ICLR emphasis:

• Reward strong empirical rigor: ablations that isolate the source of gains, multiple seeds, error bars; representation/scaling analysis where relevant.
• Keep the 9-page main text focused; move proofs, full hyperparameters, and extra results to the (unlimited) appendix.
• Include a clear reproducibility statement.

Rebuttal phase

• Plan time after notification for the discussion period.
• Prepare extra experiments reviewers are likely to ask for.
• Answer each reviewer point directly and concisely; update the PDF where allowed.

Submission checklist (ICLR)

• Official ICLR single-column style, unmodified.
• Main text within page limit; references/appendix unlimited.
• Fully anonymized (no names, no identifying repo URLs).
• Reproducibility statement included; anonymized code if possible.
• LLM-usage disclosure if the CFP requires it.
• Error bars / multiple seeds reported.
• Citations all verified (no hallucinated entries).
• No TODO/figure placeholders left; figures vector and grayscale-legible.
• OpenReview metadata, primary area, conflicts complete.
• 5-dimension self-review passed (references/writing-and-polishing.md).

When helping the user

• Ask whether the contribution is methodological, empirical, or theoretical, and what baselines exist.
• Offer to scaffold the ICLR LaTeX skeleton, draft contributions, organize appendix vs main text, build figures as placeholders, or draft rebuttal responses.