Skip to content

Evaluation

OpenAI-MRCR (Multi-Round Coreference)

Multi-needle disambiguation in multi-turn synthetic conversations

Core capabilities OpenAI

How it was run

Open a release to see the setup, scoring, and other details that the lab published. Only fields the lab actually disclosed are shown.

GPT-4.1
OpenAI

partial detail 8 method fields published

See how this test was run