Hallucination Detection on NQ_Open
79.35macF1SAPLMA + LaaB
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SAPLMA + LaaBLLM=Qwen-2.5-32B-Instruct2026.05 | 79.35 | 80.65 | |
| Lookback Lens + LaaBLLM=Qwen-2.5-32B-Instruct2026.05 | 78.08 | 79.32 | |
| SAPLMALLM=Qwen-2.5-32B-Instruct2026.05 | 76.81 | 77.55 | |
| Lookback LensLLM=Qwen-2.5-32B-Instruct2026.05 | 76.62 | 77.99 | |
| SelfCheckGPTLLM=Qwen-2.5-32B-Instruct2026.05 | 74.83 | 76.81 | |
| SAPLMA + LaaBLLM=Mistral-7B-Instruct-v0.32026.05 | 74.48 | 75.11 | |
| SAPLMA + LaaBLLM=Llama-3.1-8B-Instruct2026.05 | 73.1 | 77.13 | |
| SAPLMALLM=Llama-3.1-8B-Instruct2026.05 | 73.01 | 75.15 | |
| SAPLMALLM=Mistral-7B-Instruct-v0.32026.05 | 72.96 | 74.36 | |
| Lookback Lens + LaaBLLM=Llama-3.1-8B-Instruct2026.05 | 72.63 | 75.46 | |
| SAPLMA + LaaBLLM=Llama-3.1-70B-Instruct2026.05 | 71.71 | 78.48 | |
| Lookback Lens + LaaBLLM=Mistral-7B-Instruct-v0.32026.05 | 71.67 | 73.01 | |
| SAPLMALLM=Llama-3.1-70B-Instruct2026.05 | 71.64 | 78.62 | |
| Lookback LensLLM=Mistral-7B-Instruct-v0.32026.05 | 71.46 | 72.26 | |
| Self-JudgeLLM=Qwen-2.5-32B-Instruct2026.05 | 71.29 | 73.99 | |
| SelfCheckGPTLLM=Llama-3.1-8B-Instruct2026.05 | 69.77 | 73.17 | |
| Self-JudgeLLM=Llama-3.1-70B-Instruct2026.05 | 68.65 | 79.21 | |
| Lookback LensLLM=Llama-3.1-8B-Instruct2026.05 | 68.6 | 70.88 | |
| SelfCheckGPTLLM=Llama-3.1-70B-Instruct2026.05 | 67.73 | 77.31 | |
| Logits Lens + LaaBLLM=Qwen-2.5-32B-Instruct2026.05 | 67.21 | 70.61 | |
| Lookback Lens + LaaBLLM=Llama-3.1-70B-Instruct2026.05 | 66.65 | 76.28 | |
| SelfCheckGPTLLM=Mistral-7B-Instruct-v0.32026.05 | 66.31 | 68.07 | |
| Logits Lens + LaaBLLM=Llama-3.1-8B-Instruct2026.05 | 65.72 | 68.9 | |
| Logits LensLLM=Qwen-2.5-32B-Instruct2026.05 | 65.02 | 66.03 | |
| Eigen-ScoreLLM=Llama-3.1-8B-Instruct2026.05 | 64.82 | 67.07 | |
| Lookback LensLLM=Llama-3.1-70B-Instruct2026.05 | 64.3 | 68.67 | |
| Logits LensLLM=Llama-3.1-8B-Instruct2026.05 | 62.63 | 63.26 | |
| Eigen-ScoreLLM=Qwen-2.5-32B-Instruct2026.05 | 61.33 | 62.63 | |
| Eigen-ScoreLLM=Llama-3.1-70B-Instruct2026.05 | 60.59 | 71.6 | |
| Eigen-ScoreLLM=Mistral-7B-Instruct-v0.32026.05 | 59.14 | 62.67 | |
| Self-JudgeLLM=Llama-3.1-8B-Instruct2026.05 | 59.12 | 59.14 | |
| Logits LensLLM=Mistral-7B-Instruct-v0.32026.05 | 58.94 | 62.82 | |
| Logits Lens + LaaBLLM=Mistral-7B-Instruct-v0.32026.05 | 58.94 | 62.07 | |
| Logits Lens + LaaBLLM=Llama-3.1-70B-Instruct2026.05 | 57.21 | 71.74 | |
| Logits LensLLM=Llama-3.1-70B-Instruct2026.05 | 54.5 | 60.03 | |
| Self-JudgeLLM=Mistral-7B-Instruct-v0.32026.05 | 46.86 | 64.81 |