Hallucination Detection on WebQSP (ACC, AUC, PCC, AVG)
73.07AccuracyLUCID
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| LUCIDFramework=Readi2026.04 | 73.07 | 79.1 | 0.4043 | 64.2 | |
| LLMFramework=Readi2026.04 | 72.99 | 62.89 | 0.297 | 55.19 | |
| TrulensFramework=Readi2026.04 | 69.69 | 76.36 | 0.401 | 62.05 | |
| ReDeEPFramework=Readi, Granularity=chunk2026.04 | 68.41 | 74.88 | 0.3125 | 58.18 | |
| SelfCk-NliFramework=Readi2026.04 | 67.65 | 69.54 | 0.3353 | 56.91 | |
| LUCIDFramework=StructGPT2026.04 | 66.34 | 72.06 | 0.2292 | 53.77 | |
| ReDeEPFramework=Readi, Granularity=token2026.04 | 65.46 | 64.87 | 0.2481 | 51.71 | |
| LLMFramework=StructGPT2026.04 | 63.98 | 72.63 | 0.1803 | 51.55 | |
| RAGAsFramework=Readi2026.04 | 63.94 | 69.49 | 0.308 | 54.74 | |
| LN-EntropyFramework=Readi2026.04 | 63.37 | 65.81 | 0.2268 | 50.62 | |
| PerplexityFramework=Readi2026.04 | 62.99 | 66.88 | 0.2368 | 51.18 | |
| SelfCheckGPT-NliFramework=StructGPT2026.04 | 62.45 | 70.62 | 0.2464 | 52.57 | |
| MetaQAFramework=Readi2026.04 | 61.75 | 55.78 | 0.1121 | 42.91 | |
| PerplexityFramework=StructGPT2026.04 | 61.75 | 63.14 | 0.1851 | 47.8 | |
| Trulens GroundednessFramework=StructGPT2026.04 | 61.34 | 68.54 | 0.1683 | 48.9 | |
| LettuceDetectFramework=Readi2026.04 | 61.08 | 43.99 | 0.3337 | 46.15 | |
| SelfCk-BERTScoreFramework=Readi2026.04 | 60.89 | 64.77 | 0.2436 | 50.01 | |
| Lexical SimilarityFramework=StructGPT2026.04 | 59.39 | 61.91 | 0.1588 | 45.73 | |
| SelfCheckGPT-BERTScoreFramework=StructGPT2026.04 | 58.69 | 65.44 | 0.2275 | 48.96 | |
| Lexical SimilarityFramework=Readi2026.04 | 58.61 | 57.46 | 0.1159 | 42.55 | |
| EigenScoreFramework=StructGPT2026.04 | 57.58 | 60.42 | 0.1618 | 44.73 | |
| LN-EntropyFramework=StructGPT2026.04 | 56.75 | 65.54 | 0.2063 | 47.64 | |
| RAGAs FaithfulnessFramework=StructGPT2026.04 | 56.33 | 62.53 | 0.2255 | 47.14 | |
| ReDeEPFramework=StructGPT, granularity=chunk2026.04 | 55.22 | 61.34 | 0.2047 | 45.68 | |
| FocusFramework=Readi2026.04 | 53.95 | 52.53 | 0.0778 | 38.09 | |
| MetaQAFramework=StructGPT2026.04 | 52.16 | 56.91 | 0.074 | 38.82 | |
| EigenScoreFramework=Readi2026.04 | 50.05 | 53.26 | 0.0726 | 36.86 | |
| ReDeEPFramework=StructGPT, granularity=token2026.04 | 48.4 | 52.62 | 0.1943 | 40.15 | |
| EnergyFramework=Readi2026.04 | 47.29 | 45.55 | -0.0549 | 29.12 | |
| EnergyFramework=StructGPT2026.04 | 44.37 | 49.36 | 0.007 | 31.48 | |
| LettuceDetectFramework=StructGPT2026.04 | 35.61 | 24.16 | 0.1212 | 23.96 | |
| FocusFramework=StructGPT2026.04 | 21.97 | 45.64 | -0.018 | 21.94 |