Hallucination Detection on WikiBio Factual Sentence-level
44.23AUC-PRSelfCheckGPT
Evaluation Results
| Method | Links | |
|---|---|---|
| SelfCheckGPT2026.05 | 44.23 | |
| DS (llama-3)Backbone=Llama-32026.05 | 42.4 | |
| DS (F2LLM)Backbone=F2LLM2026.05 | 42 | |
| DS (avg)Backbone=average2026.05 | 39.5 | |
| DS (Mistral)Backbone=Mistral2026.05 | 38.1 | |
| DS (Qwen3)Backbone=Qwen32026.05 | 37.6 | |
| DS (Jina-v5)Backbone=Jina-v52026.05 | 37.2 | |
| Random2026.05 | 27 |