List your company

Multi-task Language Understanding on MMLU-Pro (Avg@8)

73.76Avg@8TTRL
68.393669.786871.1872.5732Jul 2, 2026

Evaluation results

MethodLinks
TTRLEval=cross2026.0773.76
N-OPSDEval=cross2026.0773.74
LMSIEval=cross2026.0773.46
IntuitorEval=cross2026.0773.3
Qwen3-4B base2026.0772.16
TTRLEval=in2026.0772.04
N-OPSDEval=in2026.0772.04
IntuitorEval=in2026.0771.91
LMSIEval=in2026.0768.6