List your company

Mathematical Reasoning on MathVerse-Plus Vision Dominant 1.0 (test)

41AccuracySpatialMath
9.817.92634.1Jan 24, 2026

Evaluation results

MethodLinks
SpatialMathBase Model=Phi-42026.0141
SFTBase Model=Phi-42026.0130
Zero-ShotBase Model=Phi-42026.0127
ICL (1-shot)Base Model=Phi-42026.0122
CoT (1-shot)Base Model=Phi-42026.0122
SFT + Data AugmentationBase Model=Phi-42026.0122
SFT + Data AugmentationBase Model=LLaVA-NeXT-34B2026.0120
CoT (1-shot)Base Model=LLaVA-NeXT-34B2026.0117
SpatialMathBase Model=LLaVA-NeXT-34B2026.0117
SFTBase Model=LLaVA-NeXT-34B2026.0115
Zero-ShotBase Model=LLaVA-NeXT-34B2026.0113
ICL (1-shot)Base Model=LLaVA-NeXT-34B2026.0111