naacl-experiments
Use when designing or auditing the experimental program of a NAACL submission — matching evidence to language-coverage claims, keeping cross-lingual comparisons budget-fair, testing on natively authored rather than translated data where the claim requires it, and reporting variance that survives reviewer probing.
Profile →npx skills add brycewang-stanford/Awesome-Journal-Skills --skill naacl-experiments --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.
# NAACL Experiments Design experiments backwards from the sentence you want the meta-review to contain. For NAACL-bound work that sentence almost always has a language scope in it, so the experimental program's first duty is to make the scope claim measurable — and its second duty is to make every comparison fair enough that no single reviewer probe collapses it. ## Match the design to the coverage claim | Claim you want to make | Minimum design that supports it | Design that fakes it | |---|---|---| | "Works for language X" | Natively authored X test data, native-speaker error review | Machine-translated English benchmark relabeled as X | | "Works across the Americas' languages" | Typologically spread sample (e.g., analytic + agglutinative + polysynthetic) | Three Romance languages standing in for a continent | | "Robust to dialectal variation" | Variety-labeled eval sets, per-variety breakdown | One standard variety plus vibes | | "Better than baseline B" | B re-run under equal tuning/compute budget, same prompts regime | B's two-year-old published number | | "Model-agnostic" | ≥3 model families, sizes reported | Two checkpoints of one family | Translationese deserves its own lin
- Match the design to the coverage claim
- Budget-fair comparison rules
- Variance and significance floor
- The probes NAACL reviewers run
- Experiment ledger
- Vignette: a dialect-identification study, probe by probe
- Audit sequence
- Output format
What does the naacl-experiments skill do?
Use when designing or auditing the experimental program of a NAACL submission — matching evidence to language-coverage claims, keeping cross-lingual comparisons budget-fair, testing on natively authored rather than translated data where the claim requires it, and reporting variance that survives reviewer probing.
How do I install it?
Run `npx skills add brycewang-stanford/Awesome-Journal-Skills --skill naacl-experiments --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.