Skip to content
Benchmaker

Glossary · Meaning and specification

Outer design axes

Four independent axes that locate an eval design: response space (closed to open), oracle strength (verifiable to judgment-based), maintenance (static to versioned dynamic), and interaction (single-turn to stateful trajectory).

Also called: four design axes

Defined in

Handbook · Agents and self-improvementReading agent benchmarksCapability, application or generalist? Read CyberGym, τ-bench, GDPval and others in unified coordinates before trusting a leaderboard.

See it in the full glossary →