List your company

Multi-task Language Understanding on MMLU (Source Lib., Unseen Lib., Delta)

75.68MMLU Source Lib. AccuracySIGMA
68.691270.505672.3274.1344Jun 18, 2026

Evaluation results

MethodLinks
SIGMABase Model=GPT-4o-mini2026.0675.6875.030.65
CARDBase Model=GPT-4o-mini2026.0675.1173.681.43
G-DesignerBase Model=GPT-4o-mini2026.0673.3270.053.27
GPTSwarmBase Model=GPT-4o-mini2026.0671.3764.846.53
LLM-DebateBase Model=GPT-4o-mini2026.0668.9663.735.23