Fact-Checking on MiniCheck (test)
81.2Balanced AccuracyGraphCheck-Llama3.3 70B
Evaluation Results
| Method | Links | |
|---|---|---|
| GraphCheck-Llama3.3 70BBackbone=Llama3.3 70B2025.02 | 81.2 | |
| GraphCheck-Qwen2.5 72BBackbone=Qwen2.5 72B2025.02 | 78.4 | |
| GraphEval2025.02 | 72.1 | |
| AlignScore2025.02 | 71.9 | |
| Llama3.3 70BBackbone=Llama3.3 70B2025.02 | 71 | |
| Qwen2.5 72BBackbone=Qwen2.5 72B2025.02 | 69.9 |