domains / families / research / sigir-experiments

sigir-experiments

separate

Use when designing or auditing the empirical program of a SIGIR paper — choosing test collections that match the claim, metric-cutoff discipline, paired significance testing with multiple-comparison correction, baseline tuning symmetry, ablations that isolate mechanisms, efficiency reporting, and LLM-era evaluation pitfalls.

When to use

• designing empirical programs for SIGIR papers
• auditing test collection lineups
• defining metric and cutoff discipline
• conducting paired significance testing
• ensuring baseline tuning symmetry
• designing ablations to isolate mechanisms
• reporting efficiency metrics
• identifying LLM-era evaluation pitfalls

Skill page

Source