List your company

Language Understanding on MMLU-Pro 80 (test)

38.68Pass@1FEST-DPO
8.852816.596424.3432.0836May 14, 2026

Evaluation results

MethodLinks
FEST-DPOzero-shot=true2026.0538.68
FEST-GRPOzero-shot=true2026.0536.32
CHORD-ϕzero-shot=true2026.0535.82
RLzero-shot=true2026.0534.81
ReLIFTzero-shot=true2026.0533.99
SRFTzero-shot=true2026.0533.54
RL-Gzero-shot=true2026.0533.18
HPTzero-shot=true2026.0532.83
LUFFYzero-shot=true2026.0531.72
Randomzero-shot=true2026.0510