General Task on ChartQA (accuracy)
0.895AccuracyQwen2.5-VL-72B-Instruct
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen2.5-VL-72B-InstructModel=Qwen2.5-VL-72B-Instruct, Training Strategy=Reference2025.10 | 0.895 | |
| Seed1.5-VLModel=Seed1.5-VL, Training Strategy=Reference2025.10 | 0.874 | |
| REWARDMAPModel=Qwen2.5-VL-7B-Instruct, Training Strategy=REWARDMAP2025.10 | 0.8724 | |
| Kimi-VL-A3B-InstructModel=Kimi-VL-A3B-Instruct, Training Strategy=Reference2025.10 | 0.8708 | |
| Qwen2.5-VL-7B-Instruct + SFT -> RLModel=Qwen2.5-VL-7B-Instruct, Training Strategy=SFT -> RL Baseline2025.10 | 0.8692 | |
| Qwen2.5-VL-7B-InstructModel=Qwen2.5-VL-7B-Instruct, Training Strategy=Base Model2025.10 | 0.8612 | |
| Qwen2.5-VL-3B-InstructModel=Qwen2.5-VL-3B-Instruct, Training Strategy=Reference2025.10 | 0.84 | |
| Qwen2.5-VL-32B-InstructModel=Qwen2.5-VL-32B-Instruct, Training Strategy=Reference2025.10 | 0.649 |