COLT Submission
SkillDocs & knowledgeUse when auditing a COLT (Conference on Learning Theory) submission for CMT readiness, the 12-page PMLR-format main body, single-PDF assembly with unlimited appendix, double-anonymous formatting, the parallel-submission ban covering journals and proceedings venues, and the ordered final-week checks before the AoE deadline.
Instructions available. Your AI can read the instructions. Execution depends on the setup they require.
Account requirements not reviewed. Check the skill instructions before use; ahel provides instructions and does not run this skill.
Add ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.
Then ask your AI: use the COLT Submission skill
What this skill tells your AI
The instructions your AI receives, as published by brycewang-stanford/awesome-journal-skills in COLT-Skills/skills/colt-submission/SKILL.md and read by ahel’s review.
Run this audit before uploading a learning-theory paper to COLT. Every number below was checked against the COLT 2026 Call for Papers on 2026-07-08; reopen the current CFP at learningtheory.org before trusting any of it in a later cycle, because COLT republishes its rules from scratch each year.
What the 2026 cycle actually required
- Platform: Microsoft CMT, not OpenReview. The COLT 2026 site pointed authors to
https://cmt3.research.microsoft.com/COLT2026. COLT has used other systems in the past, so never assume the portal — follow the link in the live CFP. - Deadline: February 4, 2026, Anywhere on Earth, for the 39th edition (June 29 to July 3, 2026, San Diego). One deadline for the whole manuscript; the 2026 CFP had no separate abstract-registration step, unlike NeurIPS-style venues. Confirm each year.
- Length: main body at most 12 PMLR-formatted pages excluding references, with no page limit on references and appendices, all submitted as one PDF.
- Anonymity: submissions must be suitable for double-anonymous review; the official
LaTeX template's
anonoption suppresses author names. - Scope: theoretical aspects of machine learning at the intersection of computer science, statistics, and applied mathematics, explicitly including theory that explains empirical phenomena.
The single-PDF discipline
Because the appendix rides inside the same PDF as the body, COLT has no separate supplement portal to forget — but also no second chance to patch a broken proof file. Treat the appendix as reviewable text, not as an attachment:
- Compile body plus appendix together and check the PDF page count where the 12-page boundary falls; references must start after the body, appendices after references.
- Nothing decision-critical may exist only in the appendix. A reviewer who stops at page 12 should already believe the theorem statements and the proof strategy.
- Anonymize the appendix with the same care as the body: acknowledgements, grant numbers, code URLs pointing at personal repositories, and PDF author metadata all leak.
Illustrative template skeleton
The exact style file name changes with the year (e.g., a colt2026 package for the
2026 cycle). Shape, not filename, is the stable part:
\documentclass[12pt]{article}
\usepackage[anon]{colt2026} % 'anon' hides authors for review; drop it at camera-ready
\usepackage{times}
\title[Short Title]{Tight Rates for a Learning Problem}
\coltauthor{\Name{Anonymous Author} \Email{anon@example.org}}
\begin{document}
\maketitle
\begin{abstract}%
One-paragraph statement of the problem, the main theorem, and the technique.
\end{abstract}
% <= 12 PMLR-formatted pages of body
% \bibliography{refs} -- references excluded from the limit
% \appendix -- full proofs; no page limit, same PDF
\end{document}
Download the real template from the current CFP; do not reuse a prior year's file, and never alter margins, font size, or spacing — layout tampering is a desk-reject risk at any PMLR venue.
Parallel-submission rules (verified for 2026)
The COLT 2026 CFP barred submissions substantially similar to papers previously published, accepted for publication, or in parallel submission to other peer-reviewed conferences with proceedings or to journals. It also barred sending a paper under COLT review to a journal until decisions are released or the paper is withdrawn.
| Situation | COLT 2026 reading | Safe action |
|---|---|---|
| Same results on arXiv | Preprints are standard in the theory community | Post freely; do not cite your arXiv version in a way that de-anonymizes |
| Under review at a journal | Forbidden overlap | Withdraw one venue before the COLT deadline |
| Extended abstract at a proceedings venue (e.g., STOC/FOCS/ALT) | Forbidden if substantially similar | Differentiate results substantially or pick one venue |
| Non-archival workshop talk | Generally tolerated in theory culture | Verify the workshop is truly non-archival, then declare if the form asks |
| Journal submission planned after COLT decision | Explicitly allowed once decisions are out | Calendar the journal submission after notification |
Desk-reject and severity triage
| Trigger | Severity | Window to repair |
|---|---|---|
| Body beyond 12 PMLR pages | Desk reject | Only before the deadline |
| Author names or acknowledgements left in | Desk reject or chair flag | Only before the deadline |
| Parallel-submission violation | Rejection at any stage, discoverable late | Withdraw the other venue now |
| Main theorem provable only via a gap-ridden appendix | Review-stage rejection | Fix during the proof-verification pass |
| Missing formal problem setup in the body | Review-stage damage | Fixable in the writing pass |
Final-week sequence for a COLT paper
- Freeze theorem and lemma statements; every later edit must be proofread against the appendix, because statement-proof drift is the classic COLT correctness kill.
- Run a full proof-verification pass with a coauthor who did not write the proofs, checking each inequality and each invocation of an external result.
- Rebuild the single PDF and verify the body/references/appendix boundary pages.
- Sweep for identity: author macros,
\thanks, metadata, repository links, and self-citation phrasing ("in our earlier work [3]" must become third person). - Enter CMT early — create the submission record days before February-style AoE cutoffs, since CMT accounts, conflict entry, and subject areas take time on first use.
- Upload, re-download, and open the PDF that CMT stores; confirm it is the intended compile and the appendix rendered.
CMT logistics worth doing early
- CMT accounts are per-email; make sure every coauthor's CMT identity uses their canonical email, since conflicts and the informed-AC anonymity model key off these records.
- Enter domain conflicts and coauthor conflicts completely — at a small venue an incomplete conflict list risks an actual conflicted reviewer, not a hypothetical.
- Subject-area selections steer reviewer assignment; pick the narrow area that names your subfield over the broad one that flatters the paper.
Cycle-volatility warnings
- Deadline, portal, template name, page count, and anonymity mechanics are re-decided by each year's program chairs. The 2026 facts above are anchors, not permanent rules.
- Whether COLT runs an Open Problems track in a given year, and its separate deadline and non-anonymous format, must be re-checked in the current CFP (待核实 for any cycle after 2026).
Output format
[COLT readiness] Ready / Needs fixes / Not ready
[Blocking checks] <CMT record / 12-page body / single PDF / anonymity / parallel submission>
[Proof-integrity risk] <statement-proof drift, unverified lemma, external-result misuse>
[Desk-reject risk] <one issue>
[Fix order] <ordered actions before the AoE deadline>
Signals
- GitHub stars
- 1k
- Forks
- 155
- Last commit
- Sep 2026
Advanced
- Item type
- skill
- Key
colt-submission- Source
- github.com/brycewang-stanford/awesome-journal-skills
github.com/brycewang-stanford/awesome-journal-skills
Related picks
Skill · k-dense-ai
The pick for LaTeXlatex-drawing-guide
Skill · brycewang-stanford
The pick for LaTeXgolden-pdf-ch
Skill · yusufkaraaslan
The pick for PDFpdf-co-automation
Skill · composio-community
The pick for PDFhandoff
Skill · mattpocock
More in Docs & knowledgecanvas-design
Skill · anthropics
More in Docs & knowledge