Fact Verification on TabFact (held-in)
85.02AccuracyGemini-2.5-Pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini-2.5-ProResolution=UNK2025.09 | 85.02 | |
| OpenAI-o4-miniResolution=UNK2025.09 | 84.7 | |
| QVQ-72B-PreviewResolution=Dyn.2025.09 | 77.68 | |
| Qwen2-VL-72BResolution=Dyn.2025.09 | 73.45 | |
| Qwen2-VL-7B-Table-R1Resolution=Dyn.2025.09 | 73.4 | |
| GPT-4oResolution=UNK2025.09 | 73.33 | |
| Qwen2-VL-7B-SFTResolution=Dyn.2025.09 | 73.2 | |
| Table-LLaVA 13BResolution=3362025.09 | 65 | |
| Qwen2-VL-2B-Table-R1Resolution=Dyn.2025.09 | 60.9 | |
| Claude-3.5-SonnetResolution=UNK2025.09 | 60.08 | |
| Table-LLaVA 7BResolution=3362025.09 | 59.85 | |
| Qwen2-VL-2B-SFTResolution=Dyn.2025.09 | 51.7 | |
| Qwen2-VL-7B-InsResolution=Dyn2025.09 | 40 | |
| Qwen2-VL-2B-GRPOResolution=Dyn.2025.09 | 22.7 | |
| Monkey 7BResolution=8962025.09 | 22.56 | |
| LLaVA-v1.5 7BResolution=3362025.09 | 18.9 | |
| DeepSeek-VL2 4.5BResolution=Dyn.2025.09 | 13.65 | |
| Qwen2-VL-7B-GRPOResolution=Dyn.2025.09 | 11.9 | |
| Qwen2-VL-2BResolution=Dyn.2025.09 | 8.9 | |
| mPLUG-Owl2 7BResolution=4482025.09 | 8.21 | |
| Qwen2-VL-7BResolution=Dyn.2025.09 | 7.4 |