Visual Reasoning on TSP
77Accuracy (Scale 12)EndoCoT
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| EndoCoTTraining Setting=Task-Specific Training2026.03 | 77 | 77 | 73 | 100 | |
| DiffThinkerTraining Setting=Task-Specific Training2026.03 | 76 | 72 | 59 | 97 | |
| DiffThinkerTraining Setting=Unified Training2026.03 | 64 | 49 | 34 | 94 | |
| EndoCoTTraining Setting=Unified Training2026.03 | 64 | 55 | 46 | 100 | |
| Qwen3-VL-8BTraining Setting=Task-Specific Training, Optimization Method=SFT2026.03 | 59 | 60 | 43 | 30 | |
| ThinkGenTraining Setting=Zero-Shot2026.03 | 0 | 0 | 0 | 0 | |
| ChronoEditTraining Setting=Zero-Shot2026.03 | 0 | 0 | 0 | 0 | |
| Qwen3-VL-8BTraining Setting=Zero-Shot2026.03 | 0 | 0 | 0 | 0 | |
| Qwen-Image-Edit-2511Training Setting=Zero-Shot2026.03 | 0 | 0 | 0 | 0 | |
| Qwen3-VL-8BTraining Setting=Task-Specific Training, Optimization Method=GRPO2026.03 | 0 | 0 | 0 | 0 |