Hallucination Detection on 80 Experiments Aggregated (test)
6.16Average RankKLE (heat)
Evaluation Results
| Method | Links | |
|---|---|---|
| KLE (heat)Multi-sample?=Yes2026.05 | 6.16 | |
| KLEMulti-sample?=Yes2026.05 | 6.16 | |
| CES (unsup.)Multi-sample?=No2026.05 | 6.29 | |
| KLE (full)Multi-sample?=Yes2026.05 | 6.65 | |
| Embed. Regr.Multi-sample?=Yes2026.05 | 6.92 | |
| CES (sup.)Multi-sample?=No2026.05 | 7.03 | |
| LN EntropyMulti-sample?=No2026.05 | 8.2 | |
| PerplexityMulti-sample?=No2026.05 | 8.2 | |
| Discrete SEMulti-sample?=Yes2026.05 | 8.61 | |
| LNE (bench)Multi-sample?=No2026.05 | 8.9 |