List your company

Massive Multitask Language Understanding on MMLU (en)

23.5AccuracyBPE
22.9823.11523.2523.385Jul 17, 2025

Evaluation results

MethodLinks
BPEmode=Zero-shot2025.0723.5
FXT λ3mode=Zero-shot2025.0723.3
BINOMIALmode=Zero-shot2025.0723