Hallucination Detection on NQOpenLike
85.74AUROCHIVE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| HIVEUnderlying D-LLM=Dream-7B-Instruct2026.04 | 85.74 | 96.6 | |
| HIVEBase LLM=Dream-7B-Instruct2026.04 | 85.74 | 96.6 | |
| HIVEUnderlying D-LLM=LLaDA-8B-Instruct2026.04 | 85.71 | 96.95 | |
| HIVEBase LLM=LLaDA-8B-Instruct2026.04 | 85.71 | 96.95 | |
| CCSUnderlying D-LLM=LLaDA-8B-Instruct2026.04 | 81.46 | 95.09 | |
| CCSBase LLM=LLaDA-8B-Instruct2026.04 | 81.46 | 95.09 | |
| CCSUnderlying D-LLM=Dream-7B-Instruct2026.04 | 81.31 | 92.98 | |
| CCSBase LLM=Dream-7B-Instruct2026.04 | 81.31 | 92.98 | |
| Semantic EntropyUnderlying D-LLM=LLaDA-8B-Instruct2026.04 | 81.24 | 94.57 | |
| Semantic EntropyBase LLM=LLaDA-8B-Instruct2026.04 | 81.24 | 94.57 | |
| TraceDetUnderlying D-LLM=Dream-7B-Instruct2026.04 | 79.47 | 94.86 | |
| TraceDetBase LLM=Dream-7B-Instruct2026.04 | 79.47 | 94.86 | |
| TraceDetUnderlying D-LLM=LLaDA-8B-Instruct2026.04 | 77.68 | 94.16 | |
| TraceDetBase LLM=LLaDA-8B-Instruct2026.04 | 77.68 | 94.16 | |
| EigenScoreBase LLM=LLaDA-8B-Instruct2026.04 | 76.62 | 93.47 | |
| Lexical SimilarityBase LLM=LLaDA-8B-Instruct2026.04 | 73.06 | 92.4 | |
| LN-EntropyBase LLM=LLaDA-8B-Instruct2026.04 | 72.2 | 92.23 | |
| Semantic EntropyUnderlying D-LLM=Dream-7B-Instruct2026.04 | 72.17 | 91.47 | |
| Semantic EntropyBase LLM=Dream-7B-Instruct2026.04 | 72.17 | 91.47 | |
| TSVBase LLM=LLaDA-8B-Instruct2026.04 | 71.86 | 92.94 | |
| TSVBase LLM=Dream-7B-Instruct2026.04 | 70.28 | 92.28 | |
| EigenScoreBase LLM=Dream-7B-Instruct2026.04 | 68.58 | 89.76 | |
| Lexical SimilarityBase LLM=Dream-7B-Instruct2026.04 | 66.21 | 88.74 | |
| LN-EntropyBase LLM=Dream-7B-Instruct2026.04 | 65.87 | 90.55 | |
| PerplexityBase LLM=Dream-7B-Instruct2026.04 | 60.13 | 87 | |
| PerplexityBase LLM=LLaDA-8B-Instruct2026.04 | 59.16 | 87.16 |