List your company

Vision-Language Reasoning on MMStar cleaned

77.59ScoreJigsaw + CARE
72.941274.148175.35576.5619Dec 16, 2025

Evaluation results

MethodLinks
Jigsaw + CARETraining Environment=Jigsaw puzzles, CARE=true, Backbone=Qwen 2.5 VL 7B, Evaluation Mode=thinking mode (think-answer)2025.1277.59
Qwen2.5 VLBackbone=Qwen 2.5 VL 7B, Evaluation Mode=thinking mode (think-answer)2025.1276.61
Mix + CL + CARETraining Environment=Mixed (Jigsaw, PatchFit, Rotation), Curriculum Learning=true, CARE=true, Backbone=Qwen 2.5 VL 7B, Evaluation Mode=thinking mode (think-answer)2025.1276.34
GRPO-CAREBackbone=Qwen 2.5 VL 7B, Evaluation Mode=thinking mode (think-answer)2025.1275.36
VisualSphinxBackbone=Qwen 2.5 VL 7B, Evaluation Mode=thinking mode (think-answer)2025.1275.27
Jigsaw + CL + CARETraining Environment=Jigsaw puzzles, Curriculum Learning=true, CARE=true, Backbone=Qwen 2.5 VL 7B, Evaluation Mode=thinking mode (think-answer)2025.1274.82
Vision-ZeroBackbone=Qwen 2.5 VL 7B, Evaluation Mode=thinking mode (think-answer)2025.1274.46
JigsawTraining Environment=Jigsaw puzzles, Backbone=Qwen 2.5 VL 7B, Evaluation Mode=thinking mode (think-answer)2025.1274.46
Visual JigsawBackbone=Qwen 2.5 VL 7B, Evaluation Mode=thinking mode (think-answer)2025.1274.11
ViCritBackbone=Qwen 2.5 VL 7B, Evaluation Mode=thinking mode (think-answer)2025.1273.12