List your company

Scientific Reasoning on MMLU-Pro (Mean@16, Pass@16)

25.96Mean Score @16ADS
5.867211.083616.321.5164Jun 21, 2026

Evaluation results

MethodLinks
ADSBackbone=Qwen2.5-Math-1.5B2026.0625.9674.73
ADSBackbone=Qwen3-4B-Base2026.0620.9663.8
GRPOBackbone=Qwen3-4B-Base2026.0620.0465.47
DOTS+RRBackbone=Qwen2.5-Math-1.5B2026.0619.1268.33
GRPOBackbone=Qwen2.5-Math-1.5B2026.0617.3769.8
ADSBackbone=Olmo3-7B-SFT2026.0613.5659.53
DOTS+RRBackbone=Qwen3-4B-Base2026.0612.7657.27
GRPOBackbone=Olmo3-7B-SFT2026.0610.8352.6
BaseBackbone=Qwen3-4B-Base2026.0610.4959.93
BaseBackbone=Qwen2.5-Math-1.5B2026.0610.3359.13
DOTS+RRBackbone=Olmo3-7B-SFT2026.069.251.87
BaseBackbone=Olmo3-7B-SFT2026.066.6443