Changelog
Release changes are part of the evidence trail.
Current release: polibench-paper-v1.0.1, generated from exports/polibench-clean-csv-2026-04-26.
Current Release
- Generated polibench-paper-v1.0.1 from immutable base polibench-paper-v1.0.0 and source export exports/polibench-clean-csv-2026-04-26.
- Added parser, suite, paper-release, and commit metadata to release versions.
- Added a truth gate requiring run, model, question, prompt hash, raw output, parsed answer, parser, scorer, benchmark, and lineage evidence before scores render.
- Added static schemas, schema manifest, and field-level data dictionary.
- Documented bill-analysis as separate future work: poliscore-bill-benchmark is future separate work, not part of this release.
- Created canonical sample from strict full-suite runs.
- Resolved 24 duplicate run-question pairs.
- Added exclusions, duplicate table, scoring config, benchmark version, human coding schema, and external anchor schema.