The ICLR 2026 review form, explained

How ICLR reviewers read your paper

ICLR's reviewer guide prescribes: summarize the claimed contribution; list strong and weak points; clearly state an initial recommendation with one or two key reasons; give supporting arguments; ask questions; optionally add feedback marked as not part of the decision. Reviewers should ask what problem is tackled, whether the approach is well motivated and placed in the literature, whether claims are supported correctly and rigorously, and whether the work is significant. Reviews are public on OpenReview.

• Constructive tone — criticize with kindness; reviews are public.
• Make the initial recommendation and its one or two key reasons explicit at the top of the weaknesses.
• As comprehensive as possible; excessively short reviews get flagged.
• State-of-the-art results are not required for a contribution.

Where the bar sits. ICLR 2026 accepted 27.4% of decided submissions. Paper Copilot means: posters ≈ 5.4, orals ≈ 6.4, rejected ≈ 4.0. Rough rule: mean ≥ 5 likely accept, ~4.4–5 borderline.

The reviewer form, field by field

In the order reviewers see it, with every score and its official wording.

1. 01

   Summary

   Free text

   Summarize what the paper claims to contribute. Be positive and constructive.

2. 02

   Soundness

   Score · 1–4

   Are the claims adequately supported, and are the results technically correct and rigorous?

   1. 4excellent
   2. 3good
   3. 2fair
   4. 1poor
3. 03

   Presentation

   Score · 1–4

   How clearly the paper communicates its ideas, methods and results.

   1. 4excellent
   2. 3good
   3. 2fair
   4. 1poor
4. 04

   Contribution

   Score · 1–4

   Does it provide new, relevant, valuable knowledge to the ICLR community? State-of-the-art results are not required.

   1. 4excellent
   2. 3good
   3. 2fair
   4. 1poor
5. 05

   Strengths

   List of points

   Strong points, as comprehensive as possible, each substantiated.

6. 06

   Weaknesses

   List of points

   Weak points, as comprehensive as possible. Lead with the one or two key reasons behind your recommendation, then supporting arguments.

7. 07

   Questions

   List of points

   Questions that would clarify your understanding or supply missing evidence. Additional feedback not part of the decision may be marked as such.

8. 08

   Flag For Ethics Review

   Single choice

   Choose one.

   • No ethics review needed.
   • Yes, Discrimination / bias / fairness concerns
   • Yes, Privacy, security and safety
   • Yes, Legal compliance (e.g., GDPR, copyright, terms of use)
   • Yes, Potentially harmful insights, methodologies and applications
   • Yes, Responsible research practice (e.g., human subjects, data release)
   • Yes, Research integrity issues (e.g., plagiarism, dual submission)
   • Yes, Other reasons
9. 09

   Rating

   Score · 0–10 (step 2)

   Initial recommendation. Clearly state accept or reject with one or two key reasons in the review text.

   Score	Meaning
   10	strong accept, should be highlighted at the conference
   8	accept, good paper
   6	marginally above the acceptance threshold
   4	marginally below the acceptance threshold
   2	reject, not good enough
   0	strong reject

10. 10

    Confidence

    Score · 1–5

    How certain you are about your assessment.

    Score	Meaning
    5	You are absolutely certain about your assessment. You are very familiar with the related work and checked the math/other details carefully.
    4	You are confident in your assessment, but not absolutely certain. It is unlikely, but not impossible, that you did not understand some parts of the submission or that you are unfamiliar with some pieces of related work.
    3	You are fairly confident in your assessment. It is possible that you did not understand some parts of the submission or that you are unfamiliar with some pieces of related work. Math/other details were not carefully checked.
    2	You are willing to defend your assessment, but it is quite likely that you did not understand the central parts of the submission or that you are unfamiliar with some pieces of related work. Math/other details were not carefully checked.
    1	You are unable to assess this paper and have alerted the ACs to seek an opinion from different reviewers.

Area chair meta-review

ICLR meta-reviews are public. Base the decision on the arguments in the reviews, not the mean score.

• Accept (Oral)
• Accept (Poster)
• Reject

ICLR author statements

Reviewers check these. Gaps are easy points to lose.

1. Ethics statement (recommended, max 1 page, outside the page limit)
2. Reproducibility statement pointing to code, proofs and experimental details (strongly encouraged)
3. LLM usage disclosure if LLMs played a significant role in ideation or writing (required; non-disclosure can lead to desk rejection)
4. Main text within 9 pages at submission (references and appendix unlimited)
5. Anonymized submission (no identifying links or acknowledgements)

Policy on LLMs in reviewing