Hallucination Detection on WebQuestions
87.67AUROCDRIFT
Evaluation results
| Method | Links | |
|---|---|---|
| DRIFTModel=Qwen-2.5-7B-Instruct, Input=answer2026.01 | 87.67 | |
| DRIFTModel=Qwen-2.5-7B-Instruct, Input=question2026.01 | 84.01 | |
| DRIFTModel=Gemma-3-4b-it, Input=answer2026.01 | 83.11 | |
| DRIFTModel=LLaMA 2 Chat 7B, Input=answer2026.01 | 82.76 | |
| DRIFTModel=LLaMA 2 Chat 7B, Input=question2026.01 | 80.79 | |
| Semantic EntropyModel=Qwen-2.5-7B-Instruct2026.01 | 80.71 | |
| HaloScopeModel=Qwen-2.5-7B-Instruct, Input=question2026.01 | 80.43 | |
| Semantic EntropyModel=LLaMA 2 Chat 7B2026.01 | 80.12 | |
| DRIFTModel=Gemma-3-4b-it, Input=question2026.01 | 79.63 | |
| HaloScopeModel=Qwen-2.5-7B-Instruct, Input=answer2026.01 | 78.56 | |
| HaloScopeModel=LLaMA 2 Chat 7B, Input=answer2026.01 | 77.03 | |
| Semantic EntropyModel=Gemma-3-4b-it2026.01 | 76.76 | |
| HaloScopeModel=Gemma-3-4b-it, Input=question2026.01 | 76.7 | |
| HaloScopeModel=Gemma-3-4b-it, Input=answer2026.01 | 75.72 | |
| HaloScopeModel=LLaMA 2 Chat 7B, Input=question2026.01 | 72.62 |