agentsclimarketplace

Ijcai topic selection

Skill brycewang-stanford/Awesome-Journal-Skills/IJCAI-Skills/skills/ijcai-topic-selection

Use when deciding whether an AI project is a strong IJCAI or IJCAI-ECAI fit, comparing main track versus special tracks, Survey Track, AIJ/JAIR expedited-publication potential, or alternative AI venues such as NeurIPS, ICML, ICLR, AAAI, AISTATS, UAI, COLT, KDD, CVPR, ACL, or ICRA.From its SKILL.md

Install
npx -y skills add brycewang-stanford/Awesome-Journal-Skills --skill ijcai-topic-selection

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

SKILL.md

3.6 KB, 737 tokens by cl100k_base, as published. Nobody here has run it

IJCAI Topic Selection

Use this before writing. IJCAI main-track submissions should report significant, original, previously unpublished AI results with broad relevance to the AI community.

Fit test

  • Ask whether the paper advances an AI problem, AI technique, AI application domain, or cross-disciplinary AI boundary rather than only reporting an application result.
  • Prefer IJCAI when the contribution is broadly AI-facing, technically complete in a compact format, and can survive a two-phase review process with limited response leverage.
  • Consider special tracks for human-centered AI, AI for social good, AI4Tech, AI and health, or AI and robotics when the venue-specific framing is stronger than a general main-track frame.
  • Consider the Survey Track only for mature areas and author teams with established topic expertise.
  • Route elsewhere if the contribution is primarily ML theory, representation learning, computer vision, NLP, robotics, HCI, systems, or data mining and a specialist venue has a clearer audience.
  • Check policy fit early: anonymity, recent rejection disclosures, dual-submission limits, data ethics, LLM-use constraints, and in-person presentation feasibility.

Venue routing matrix

IJCAI's edge is breadth and a compact format, but a sharply specialist contribution often finds a clearer audience elsewhere. Route by where the strongest reviewers and framing sit.

Contribution profileStrong IJCAI fitConsider instead
Cross-cutting AI idea (search, KR, planning, multi-agent) readable across subfieldsYes, main track
Deep ML theory or representation learningSometimesNeurIPS, ICML, COLT
Heavy CV or NLP systemRarely the best homeCVPR, ACL
Mature area survey by established authorsSurvey TrackJournal survey
Polished, complete result wanting fast journal routeMain track plus AIJ/JAIR ecosystem
Social-good or human-centered AI framingSpecial trackDomain venue

Worked vignette: a game-theory mechanism

A mechanism-design result for multi-agent resource allocation could go to an economics-and- computation venue or to IJCAI. If the contribution is an AI technique broadly relevant across agents, search, and learning audiences and fits a compact 7-page body, IJCAI's main track is a strong home; the game-theory/economic-paradigms scope is squarely in-bounds. If the core advance is a fine-grained economic-theory proof aimed at a specialist EC audience, route there instead. Decide by which reviewer pool best judges the central claim.

Reviewer/fit pushback and the venue-specific fix

  • "This is an application, not an AI advance." Reframe the AI technique or problem contribution, or route to a domain venue.
  • "Too specialized for a broad PC." Either broaden the framing for IJCAI breadth or choose the specialist venue with a clearer audience.
  • "Survey from a junior team." Reserve the Survey Track for mature areas and established topic expertise.

Output format

[Fit] strong IJCAI / possible IJCAI / better elsewhere
[Best track] main / special / survey / other venue
[Contribution sentence] <one sentence>
[Top rejection risk] <novelty/clarity/evidence/policy/scope>
[Next action] <experiment, framing, related work, or venue switch>

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.