Conceptual Reasoning

Tests whether model judgments rank argumentative critiques in the same order as expert human ratings across philosophy, AI alignment, and other concept-heavy texts.

Released

Unknown

Readiness

Paper only

Primary field

General AI

Benchmark Radar records only publicly supported details and links back to primary sources for verification.