Fact-checking on LIAR-RAW
77.38PrecisionKG-CRAFT
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| KG-CRAFTBackbone=Llama 3.32026.01 | 77.38 | 70.67 | 73.87 | |
| KG-CRAFTBackbone=Claude 3.72026.01 | 73.92 | 70.86 | 72.36 | |
| KG-CRAFTBackbone=Claude 3.52026.01 | 62.7 | 59.38 | 60.99 | |
| Llama 3.3 70B2026.01 | 47.21 | 23.52 | 31.4 | |
| DelphiAgentBackbone=gpt-4o-mini2026.01 | 32.79 | 22.33 | 26.03 | |
| FactLLAMAknow2026.01 | 32.46 | 32.05 | 30.44 | |
| FactLLAMA2026.01 | 32.32 | 31.57 | 29.98 | |
| L-DefenseBackbone=LLAMA22026.01 | 31.63 | 31.71 | 31.4 | |
| DelphiAgentBackbone=gpt-4o2026.01 | 31.33 | 28.36 | 28.36 | |
| L-DefenseBackbone=ChatGPT2026.01 | 30.55 | 32.2 | 30.53 | |
| EExpFND2026.01 | 29.91 | 29.93 | 29.58 | |
| CofCED2026.01 | 29.48 | 29.55 | 28.93 | |
| Claude 3.5 Sonnet2026.01 | 29.25 | 27.83 | 28.52 | |
| Claude 3.7 Sonnet2026.01 | 28.26 | 26.63 | 27.42 | |
| GenFE2026.01 | 28.01 | 26.16 | 26.49 | |
| ChatGPT 3.5 Turbo2026.01 | 25.41 | 27.33 | 25.11 | |
| SBERT-FC2026.01 | 24.09 | 22.07 | 22.19 | |
| dEFEND2026.01 | 23.09 | 18.56 | 17.51 | |
| GenFE-MT2026.01 | 18.55 | 19.9 | 15.15 | |
| Llama 2 7B2026.01 | 17.11 | 17.37 | 15.14 |