AI BENCHMARK PROFILE

MRCR 1M (pointwise)

General AILong Context & Memory

MRCR 1M (pointwise) is a variant of the Multi-Round Coreference Resolution benchmark that uses pointwise evaluation for ultra-long contexts (~1M tokens). This version evaluates each response independently rather than comparatively, testing models' absolute performance on long-context reasoning tasks.

Released: Unknown
Readiness: Paper only
Primary field: General AI

Primary resources

Paper ↗

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