Table Fact Verification on TabFact small
85.77Overall AccuracyTab-PoT
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Tab-PoTLLM=DeepSeek-67B2024.06 | 85.77 | 91.34 | 80.27 | |
| DaterLLM=Codex2024.06 | 85.6 | 91.2 | 80 | |
| BinderLLM=Codex2024.06 | 85.1 | — | — | |
| COTLLM=DeepSeek-67B2024.06 | 82.61 | 88.46 | 76.84 | |
| Tab-PoTLLM=Mixtral-8x7B2024.06 | 81.67 | 88.36 | 75.07 | |
| COTLLM=Mixtral-8x7B2024.06 | 80.09 | 86.07 | 74.19 | |
| DirectLLM=DeepSeek-67B2024.06 | 78.61 | 84.68 | 72.62 | |
| DirectLLM=Mixtral-8x7B2024.06 | 75.89 | 81.29 | 70.56 | |
| POTLLM=DeepSeek-67B2024.06 | 74.51 | 76.32 | 72.72 | |
| POTLLM=Mixtral-8x7B2024.06 | 73.32 | 76.02 | 70.66 | |
| Tab-PoTLLM=Mixtral-7B2024.06 | 71.84 | 76.82 | 66.93 | |
| COTLLM=Mixtral-7B2024.06 | 70.41 | 73.73 | 67.12 | |
| DirectLLM=Mixtral-7B2024.06 | 67.89 | 73.13 | 62.71 | |
| POTLLM=Mixtral-7B2024.06 | 66.25 | 65.97 | 66.54 | |
| Tab-PoTLLM=DeepSeek-7B2024.06 | 65.86 | 70.75 | 61.04 | |
| COTLLM=DeepSeek-7B2024.06 | 65.17 | 70.65 | 59.76 | |
| POTLLM=DeepSeek-7B2024.06 | 61.91 | 64.88 | 58.98 | |
| DirectLLM=DeepSeek-7B2024.06 | 57.71 | 59.5 | 55.94 |