Hallucination Detection on Dolly Llama2-7B (test)
77.78AccRAGLens
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| RAGLens2025.12 | 77.78 | 87.64 | 80.7 | |
| ReDeEP2025.12 | 73.73 | 79.49 | 78.33 | |
| TurLens2025.12 | 68 | 62.64 | 65.67 | |
| EigenScore2025.12 | 65.96 | 67.86 | 72.41 | |
| RAGAS2025.12 | 65.6 | 66.48 | 63.92 | |
| LwMLM2025.12 | 65.5 | — | 77.02 | |
| Perplexity2025.12 | 63.63 | 68.25 | 70.97 | |
| Focus2025.12 | 62.12 | 67.83 | 65.45 | |
| Prompt2025.12 | 62 | — | 54.76 | |
| ChainPoll2025.12 | 62 | 65.93 | 55.81 | |
| RefCheck2025.12 | 61 | 64.94 | 54.12 | |
| SAPLMA2025.12 | 60.84 | 65 | 66.53 | |
| SEP2025.12 | 60.6 | 60.67 | 70.23 | |
| Llama2-13B(LR)classifier=Logistic Regression, features=Llama2-13B2025.12 | 60.43 | — | 66.16 | |
| ITI2025.12 | 58 | 54.94 | 62.81 | |
| FactCC2025.12 | 57.58 | 61.69 | 58.82 | |
| LN-Entropy2025.12 | 56.56 | 60.74 | 62.61 | |
| Energy2025.12 | 56.56 | 60.74 | 62.61 | |
| FAcTScore2025.12 | 53.54 | 48.13 | 68.49 | |
| P(True)2025.12 | 53.44 | 61.91 | 50.95 | |
| SelfCheckGPT2025.12 | 53 | — | 31.88 |