Table Question Answering on VWTQ_syn (test)
72.8AccuracyGPT-4o
Evaluation results
| Method | Links | |
|---|---|---|
| GPT-4oModel Scale=Large Scale2026.06 | 72.8 | |
| VTool-R1 (GRPO)Model Scale=Large Scale, Backbone=Qwen-2.5VL-7B2026.06 | 68.4 | |
| IAPOModel Scale=Small Scale, Backbone=Qwen-2.5VL-3B2026.06 | 60.2 | |
| ToRLModel Scale=Small Scale, Backbone=Qwen-2.5VL-3B2026.06 | 56.1 | |
| TRMModel Scale=Small Scale, Backbone=Qwen-2.5VL-3B2026.06 | 54.6 | |
| Qwen-2.5VL-7BModel Scale=Large Scale2026.06 | 52.6 | |
| VTool-R1 (GRPO)Model Scale=Small Scale, Backbone=Qwen-2.5VL-3B2026.06 | 52.6 | |
| Qwen-2.5VL-3BModel Scale=Small Scale2026.06 | 26.3 |