Share your thoughts, 1 month free Claude Pro on usSee more

Video Reasoning on VideoMathQA

55.76Accuracy

CoPD

Updated 1mo ago

Evaluation Results

Method	Links
CoPD 2026.04		55.76
Mixed RLVR 2026.04		55.71
MOPD 2026.04		53.43
Image-Expert 2026.04		53.1
Video-Expert 2026.04		50.95
Base 2026.04		49.33
Text-Expert 2026.04		47.19
Video-Zero 2026.05		32.62
Video-Zero 2026.05		32.62
ProcessThinker (process-only) 2026.04		31.67
Video-Zero 2026.05		30.48
MiMo-SFT-7B 2026.05		29.76
Video-Zero 2026.05		29.52
Video-Zero 2026.05		27.86
ProcessThinker (outcome-only) 2026.04		27.86
ProcessThinker (outcome + process) 2026.04		27.86
VIDEO-R1-7B 2026.04		26.67
Video-Zero 2026.05		26.19
Video-Zero (Peak) 2026.05		26.19
Video-Zero 2026.05		25.95
Video-Zero 2026.05		25.48
Video-Zero (Iter 4) 2026.05		25.48
Video-Zero 2026.05		25.24
QWEN3-VL-8B-INSTRUCT 2026.04		25.2
VisPlay 2026.05		24.76
V-Zero 2026.05		24.52
V-Zero 2026.05		24.05
Base Model 2026.05		23.81
Base Model 2026.05		23.81
Video-Zero 2026.05		23.81
PROCESSTHINKER-SFT 2026.04		23.57
VisPlay 2026.05		23.33
Video-Zero 2026.05		23.33
Video-Zero 2026.05		23.33
Video-Zero 2026.05		23.1
V-Zero 2026.05		22.86
Video-Zero 2026.05		22.62
Video-Zero (Peak) 2026.05		22.62
VisPlay 2026.05		22.38
V-Zero 2026.05		22.38
V-Zero 2026.05		22.14
V-Zero 2026.05		22.14
Video-Zero 2026.05		21.9
Video-Zero 2026.05		21.67
VisPlay 2026.05		21.67
Video-Zero (Iter 4) 2026.05		21.67
VisPlay 2026.05		21.43
VisPlay 2026.05		21.43
Video-Zero 2026.05		21.19
V-Zero 2026.05		20.95
V-Zero 2026.05		20.95
VisPlay 2026.05		20.71
VisPlay 2026.05		20.71
V-Zero 2026.05		20.48
V-Zero 2026.05		20.48
V-Zero 2026.05		20.48
Video-Zero 2026.05		19.76
VisPlay 2026.05		19.52
V-Zero 2026.05		19.52
Base Model 2026.05		19.05
Video-Zero 2026.05		19.05
Base Model 2026.05		19.05
VisPlay 2026.05		18.81
VisPlay 2026.05		18.33
VisPlay 2026.05		18.33
Video-Zero 2026.05		17.86
InternVL3.5-4B 2026.05		12.62