Skip to content

Glossary · Reliability, calibration, and statistics

Linking items

Tasks carried unchanged across versions of a refreshed eval suite, with overlap samples, so that scores on different versions can be linked. Some sources call them anchor tasks; this handbook reserves 'anchor' for reference anchors and the anchor channel.

Also called: anchor tasks, linking set

Defined in

Handbook · LifecycleLiving eval suitesThe asset is the machinery that notices when a score stops meaning what it used to: tripwires, drift and saturation.
Reference anchor
The external object a judgment is rooted in: a canonical answer, gold state, reference distribution, adjudicated expert labels, a stronger evaluator, or production outcomes. Gold is a reference anchor, not a mechanism or a rung.
Versioned dynamism
Dynamic corpus, static run: refresh an eval suite deliberately across versions, but freeze every environment snapshot, tool contract, scorer, and dependency needed to reproduce a reported result. A new snapshot is a new measurement form.

Where it comes up

See it in the full glossary →