Question Answering on WTQ (held-in)
80.34AccuracyGemini-2.5-Pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini-2.5-ProResolution=UNK2025.09 | 80.34 | |
| OpenAI-o4-miniResolution=UNK2025.09 | 78.2 | |
| Claude-3.5-SonnetResolution=UNK2025.09 | 71.8 | |
| QVQ-72B-PreviewResolution=Dyn.2025.09 | 68.2 | |
| Qwen2-VL-72BResolution=Dyn.2025.09 | 65.7 | |
| GPT-4oResolution=UNK2025.09 | 64.39 | |
| Qwen2-VL-7B-GRPOResolution=Dyn.2025.09 | 53.2 | |
| Qwen2-VL-7B-Table-R1Resolution=Dyn.2025.09 | 50.3 | |
| Qwen2-VL-7B-SFTResolution=Dyn.2025.09 | 46.8 | |
| Qwen2-VL-7BResolution=Dyn.2025.09 | 46.5 | |
| DeepSeek-VL2 4.5BResolution=Dyn.2025.09 | 40.42 | |
| Qwen2-VL-2B-GRPOResolution=Dyn.2025.09 | 35.3 | |
| Qwen2-VL-2B-Table-R1Resolution=Dyn.2025.09 | 34.4 | |
| Qwen2-VL-2B-SFTResolution=Dyn.2025.09 | 31 | |
| Qwen2-VL-2BResolution=Dyn.2025.09 | 22.3 | |
| Table-LLaVA 13BResolution=3362025.09 | 20.41 | |
| Qwen2-VL-7B-InsResolution=Dyn2025.09 | 19.73 | |
| Monkey 7BResolution=8962025.09 | 19.07 | |
| Table-LLaVA 7BResolution=3362025.09 | 18.43 | |
| LLaVA-v1.5 7BResolution=3362025.09 | 1.24 | |
| mPLUG-Owl2 7BResolution=4482025.09 | 0.67 |