Hallucination Detection on AG News
80.2PRRUQ_ICL
Evaluation Results
| Method | Links | |
|---|---|---|
| UQ_ICLModel=LLaMA2-13B2026.04 | 80.2 | |
| Function VectorModel=LLaMA2-7B2026.04 | 78.5 | |
| Function VectorModel=LLaMA2-13B2026.04 | 78.3 | |
| Self-FVModel=LLaMA2-7B2026.04 | 77.9 | |
| Self-FVModel=LLaMA2-13B2026.04 | 77.7 | |
| Total EntropyModel=LLaMA2-7B2026.04 | 77.6 | |
| Total EntropyModel=LLaMA2-13B2026.04 | 75.8 | |
| UQ_ICLModel=LLaMA2-7B2026.04 | 55.6 | |
| Semantic EntropyModel=LLaMA2-7B2026.04 | 54.8 | |
| Semantic EntropyModel=LLaMA2-13B2026.04 | 53.6 |