List your company

Scientific Reasoning (Physics, Chemistry, Biology) on MMK-12

61.9Physics AccuracySTRIDE
44.7449.19553.6558.105Jun 14, 2026

Evaluation results

MethodLinks
STRIDEBase model=Qwen2.5-VL-7B2026.0661.971.572.3
SRPOBase model=Qwen2.5-VL-7B2026.0660.670.469.5
MM-EurekaBase model=Qwen2.5-VL-7B2026.0656.265.265.2
BaseBase model=Qwen2.5-VL-7B2026.0645.456.454