Chart Question Answering on ChartQA Vertical (test)
81.7AccuracyGPT-4o
Evaluation results
| Method | Links | |
|---|---|---|
| GPT-4oModel Scale=Large Scale2026.06 | 81.7 | |
| VTool-R1 (GRPO)Model Scale=Large Scale, Backbone=Qwen-2.5VL-7B2026.06 | 75.4 | |
| IAPOModel Scale=Small Scale, Backbone=Qwen-2.5VL-3B2026.06 | 64.1 | |
| ToRLModel Scale=Small Scale, Backbone=Qwen-2.5VL-3B2026.06 | 63.1 | |
| TRMModel Scale=Small Scale, Backbone=Qwen-2.5VL-3B2026.06 | 62.7 | |
| VTool-R1 (GRPO)Model Scale=Small Scale, Backbone=Qwen-2.5VL-3B2026.06 | 60.9 | |
| Qwen-2.5VL-7BModel Scale=Large Scale2026.06 | 48.7 | |
| Qwen-2.5VL-3BModel Scale=Small Scale2026.06 | 22.8 |