foresight
A historical evaluation of the scoring model against known outcomes. Returns hit rate, precision@k, calibration curve.
A Backtest Run replays a frozen prediction batch against the outcomes the world has since revealed. The result is a hit rate, a precision@k, a calibration curve, and a per-paper accuracy breakdown — the receipts that let outside readers judge whether Foresight is well-calibrated.
Source: curated glossary catalog. Freshness: git_versioned_curated_catalog.
Term API · apps/web/data/glossary/terms.ts