Hallucination Detection on QALD English 10
75.37AccuracyLUCID
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| LUCIDFramework=StructGPT2026.04 | 75.37 | 81.22 | 0.4255 | 66.38 | |
| ReDeEPFramework=StructGPT, granularity=chunk2026.04 | 74.63 | 80.27 | 0.4178 | 65.56 | |
| LLMFramework=StructGPT2026.04 | 74.33 | 59.87 | 0.3159 | 55.26 | |
| SelfCheckGPT-NliFramework=StructGPT2026.04 | 72.01 | 81.29 | 0.4218 | 65.16 | |
| LUCIDFramework=Readi2026.04 | 71.71 | 76.16 | 0.3629 | 61.39 | |
| SelfCheckGPT-BERTScoreFramework=StructGPT2026.04 | 70.15 | 75.27 | 0.4106 | 62.16 | |
| ReDeEPFramework=Readi, Granularity=chunk2026.04 | 69.76 | 75.29 | 0.3466 | 59.9 | |
| LLMFramework=Readi2026.04 | 69.23 | 67.79 | 0.348 | 57.27 | |
| Lexical SimilarityFramework=StructGPT2026.04 | 69.03 | 69.36 | 0.3344 | 57.28 | |
| SelfCk-NliFramework=Readi2026.04 | 68.78 | 72.46 | 0.3469 | 58.64 | |
| PerplexityFramework=StructGPT2026.04 | 68.66 | 73.64 | 0.3805 | 60.12 | |
| LN-EntropyFramework=StructGPT2026.04 | 68.66 | 72.04 | 0.3417 | 58.29 | |
| LettuceDetectFramework=StructGPT2026.04 | 67.31 | 60.35 | 0.2434 | 50.67 | |
| FocusFramework=StructGPT2026.04 | 64.55 | 54.32 | 0.1599 | 44.95 | |
| TrulensFramework=Readi2026.04 | 64.04 | 67.78 | 0.263 | 52.71 | |
| PerplexityFramework=Readi2026.04 | 63.9 | 66.24 | 0.2462 | 51.59 | |
| EigenScoreFramework=StructGPT2026.04 | 62.31 | 63.78 | 0.2211 | 49.4 | |
| Trulens GroundednessFramework=StructGPT2026.04 | 62.17 | 72.25 | 0.3973 | 58.05 | |
| Lexical SimilarityFramework=Readi2026.04 | 60.98 | 62.99 | 0.1898 | 47.65 | |
| SelfCk-BERTScoreFramework=Readi2026.04 | 60.98 | 66.91 | 0.2934 | 52.41 | |
| ReDeEPFramework=Readi, Granularity=token2026.04 | 60.98 | 65.83 | 0.3072 | 52.51 | |
| ReDeEPFramework=StructGPT, granularity=token2026.04 | 60.45 | 64.63 | 0.3989 | 54.99 | |
| LN-EntropyFramework=Readi2026.04 | 60 | 65.82 | 0.2117 | 49 | |
| MetaQAFramework=Readi2026.04 | 60 | 65.58 | 0.2117 | 48.92 | |
| RAGAs FaithfulnessFramework=StructGPT2026.04 | 59.33 | 65.84 | 0.2741 | 50.86 | |
| MetaQAFramework=StructGPT2026.04 | 58.58 | 62.93 | 0.2598 | 49.16 | |
| FocusFramework=Readi2026.04 | 56.1 | 48.37 | 0.0618 | 36.88 | |
| RAGAsFramework=Readi2026.04 | 56.1 | 61.79 | 0.2013 | 46.01 | |
| EigenScoreFramework=Readi2026.04 | 54.63 | 54.47 | 0.1412 | 41.07 | |
| EnergyFramework=Readi2026.04 | 50.24 | 44.16 | -0.0186 | 30.85 | |
| EnergyFramework=StructGPT2026.04 | 49.63 | 46.58 | -0.0411 | 30.7 | |
| LettuceDetectFramework=Readi2026.04 | 48.29 | 40 | 0.2724 | 38.51 |