Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →brycewang-stanford/cikm-supplementary
Use when deciding what supporting material accompanies a CIKM submission given budgets that count appendices inside the page limit, structuring the in-PDF appendix versus the anonymously cited artifact, keeping both double-blind, and handling the uncounted GenAI-disclosure and reference sections correctly.
Passed 22 of 25 test cases with the skill, and 10 of 25 without it.
and 25 more in the full report
15 fixed7 passed either way0 still failing3 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 15 test cases the skill fixed. Same model, same prompt, with and without the skill.
In our CIKM full paper draft, our references currently take up 3 pages following our 10-page core paper. Does CIKM allow 3 pages of references for full papers?
| Checked | Without skill | With skill |
|---|---|---|
| States that references for CIKM full track papers are capped at up to 2 pages outside the page budget. | Missed | Passed |
| Identifies that 3 pages of references exceed the allowable reference page limit. | Missed | Passed |
The instructions your model reads when the skill loads.
CIKM 2026 gives supporting material no free rider: every track's budget counts the appendix inside the page limit — 10 pages for full papers, 4/7/4/4 for short/applied/resource/demo — and only the references (full track: up to 2 pages) plus the GenAI Usage Disclosure section sit outside it (source map, 2026-07-08). There is no separate supplementary-upload channel in the verified 2026 calls. The design question is therefore three-way: costed appendix pages, the anonymously cited artifact, or cut.
| Container | Costs | Right contents |
|---|---|---|
| Main body | Budget pages | Everything needed to evaluate the claim unaided |
| In-PDF appendix | Budget pages too | Only material reviewers must see during review that the body cannot hold: a proof, a prompt set, a schema diagram |
| Cited anonymous artifact | Zero pages, but discretionary attention | Configs, full hyperparameter grids, extra per-dataset tables, code, data cards |
Decision rule per candidate item: must a reviewer see this to score the paper? If yes and it is short, appendix — knowing it displaces body content one-for-one. If yes and it is long, restructure the body so a summary suffices, and put the whole in the artifact. If no, artifact or delete.
/home/username/ — the classic leak channels.The GenAI Usage Disclosure section and the references are outside the budget — and
test cases passed, vs 10 / 25 without
Published skills whose descriptions closely match this one, with their headline benchmark lift.
More skills from the same source and nearby in the registry, with their headline benchmark lift.