Hallucination Detection on Emotion
41PRRFunction Vector
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Function VectorModel=LLaMA2-13B2026.04 | 41 | — | |
| Self-FVModel=LLaMA2-13B2026.04 | 39.4 | — | |
| Total EntropyModel=LLaMA2-13B2026.04 | 38.6 | — | |
| Function VectorModel=LLaMA2-7B2026.04 | 37.4 | — | |
| Self-FVModel=LLaMA2-7B2026.04 | 36.8 | — | |
| Total EntropyModel=LLaMA2-7B2026.04 | 36.4 | — | |
| Semantic EntropyModel=LLaMA2-13B2026.04 | 36.4 | — | |
| UQ_ICLModel=LLaMA2-7B2026.04 | 35.6 | — | |
| Semantic EntropyModel=LLaMA2-7B2026.04 | 28.3 | — | |
| UQ_ICLModel=LLaMA2-13B2026.04 | 3.6 | — | |
| Function VectorModel=LLaMA2-7B2026.04 | — | 65.1 | |
| Function VectorModel=LLaMA2-13B2026.04 | — | 67.6 | |
| Self-FVModel=LLaMA2-7B2026.04 | — | 64.7 | |
| Self-FVModel=LLaMA2-13B2026.04 | — | 66.8 | |
| Semantic EntropyModel=LLaMA2-7B2026.04 | — | 53.8 | |
| Semantic EntropyModel=LLaMA2-13B2026.04 | — | 48.9 | |
| Total EntropyModel=LLaMA2-7B2026.04 | — | 64.6 | |
| Total EntropyModel=LLaMA2-13B2026.04 | — | 66.6 | |
| UQ_ICLModel=LLaMA2-7B2026.04 | — | 68.1 | |
| UQ_ICLModel=LLaMA2-13B2026.04 | — | 68 |