List your company

Mathematical Reasoning on MathVerse (test)

44.88AccuracyQwen2.5-VL-7B + Cont. Reward
41.988842.739443.4944.2406Nov 20, 2025

Evaluation results

MethodLinks
Qwen2.5-VL-7B + Cont. Rewardreward_type=Continuous2025.1144.88
Vision-Zeroexternal supervision=true2025.1143.86
Qwen2.5-VL-7B (Baseline)2025.1143.78
Qwen2.5-VL-7B + Discrete Rewardreward_type=Discrete2025.1142.1