Fact-checking on AggreFact CNN
69.5Balanced AccGraphEval
Evaluation Results
| Method | Links | |
|---|---|---|
| GraphEval2025.02 | 69.5 | |
| Claude 3.5-SonnetScale=>300B2025.02 | 68.8 | |
| GPT-4oScale=>300B2025.02 | 66.8 | |
| GraphCheckBackbone=Qwen 72B2025.02 | 66.5 | |
| OpenAI o1Scale=>300B2025.02 | 65.3 | |
| MiniCheck2025.02 | 63.7 | |
| DeepSeek-V3 671BScale=>300B2025.02 | 63.2 | |
| GraphCheckBackbone=Llama3.3 70B2025.02 | 62.4 | |
| GPT-4Scale=>300B2025.02 | 60.7 | |
| AlignScore2025.02 | 54.1 | |
| Llama3.3 70B2025.02 | 53.5 | |
| Llama3 8B2025.02 | 51.3 | |
| ACUEval2025.02 | 50 | |
| Qwen2.5 72B2025.02 | 49.9 | |
| Qwen2.5 7B2025.02 | 45.3 |