List your company

Massive Multitask Language Understanding on MMLU hi

24.8MMLU hi AccuracyFXT λ3
24.2824.41524.5524.685Jul 17, 2025

Evaluation results

MethodLinks
FXT λ3mode=Zero-shot2025.0724.8
BPEmode=Zero-shot2025.0724.3
BINOMIALmode=Zero-shot2025.0724.3