uai-artifact-evaluation
Use when packaging code, data, samplers, solvers, and logs for a UAI submission's 50 MB supplementary ZIP or a public post-acceptance release, making probabilistic-inference claims independently runnable while keeping every file double-blind, given that UAI reviewers may open the archive but are not obliged to read it.
Profile →npx skills add brycewang-stanford/Awesome-Journal-Skills --skill uai-artifact-evaluation --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
# UAI Artifact Evaluation UAI posted no separate artifact-evaluation track or badge system for the 2026 cycle (existence of one in later cycles: 待核实). What it did post is sharper than a badge: code and data release is strongly encouraged, a 50 MB anonymous ZIP rides with the submission, and reviewers judge whether claims are "backed up convincingly." The artifact's job is to convert a skeptical probabilistic-ML reviewer's spot-check into confirmation. ## Design for the five-minute skeptic Because reviewers are explicitly not required to consult supplementary material, assume whoever opens the ZIP gives it five minutes. Optimize for that reader: - A top-level `README` that states, in its first ten lines, what claim each script reproduces and how long it takes. - One command per headline result, each with a small-scale mode that finishes on a laptop: a reviewer who reproduces Figure 2 at n=500 in ninety seconds will extend you trust for the n=50,000 version. - Pinned dependencies and one canonical entry point. A `Makefile` or single driver script beats a directory of loose notebooks. - Expected outputs stored beside the scripts, so "did it work?" needs no judgment. ## What inference
- Design for the five-minute skeptic
- What inference artifacts must expose
- Anonymity inside the archive
- README skeleton for the archive
- Data that cannot ship
- Licensing notes
- After acceptance
- Output format
Build an anonymous artifact from a clean tree, never from the working repo
git archive --format=tar HEAD | tar -x -C /tmp/uai-artifact
cd /tmp/uai-artifact
grep -rniE 'university|\.edu|author|thanks|grant' --include='*.py' --include='*.md' . | head
find . -name '*.ipynb' -exec grep -l '"authors"' {} \; # notebook metadata
rm -rf .git .github; find . -name '.DS_Store' -delete
zip -r ../supplement.zip . && du -h ../supplement.zip # must be ≤ 50 MB (2026 cap)What does the uai-artifact-evaluation skill do?
Use when packaging code, data, samplers, solvers, and logs for a UAI submission's 50 MB supplementary ZIP or a public post-acceptance release, making probabilistic-inference claims independently runnable while keeping every file double-blind, given that UAI reviewers may open the archive but are not obliged to read it.
How do I install it?
Run `npx skills add brycewang-stanford/Awesome-Journal-Skills --skill uai-artifact-evaluation --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From brycewang-stanford/Awesome-Journal-Skills, a repository with 909 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.