Hallucination Detection on CNN/DailyMail
0.694AvgWDAvgWD
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| AvgWDModel=Qwen2-72b2026.03 | 0.694 | — | — | — | — | — | — | |
| AvgWDModel=Qwen2-7b2026.03 | 0.679 | — | — | — | — | — | — | |
| AvgWDModel=Llama-2-7b2026.03 | 0.666 | — | — | — | — | — | — | |
| AvgWDModel=Llama-3.2-3b2026.03 | 0.658 | — | — | — | — | — | — | |
| AvgWDModel=Llama-3.1-8b2026.03 | 0.625 | — | — | — | — | — | — | |
| DSEModel=Llama-3.2-3b2026.03 | — | — | — | — | 0.628 | — | — | |
| DSEModel=Llama-2-7b2026.03 | — | — | — | — | 0.65 | — | — | |
| DSEModel=Llama-3.1-8b2026.03 | — | — | — | — | 0.615 | — | — | |
| DSEModel=Qwen2-7b2026.03 | — | — | — | — | 0.654 | — | — | |
| DSEModel=Qwen2-72b2026.03 | — | — | — | — | 0.658 | — | — | |
| EigenWDModel=Llama-3.2-3b2026.03 | — | 0.674 | — | — | — | — | — | |
| EigenWDModel=Llama-2-7b2026.03 | — | 0.668 | — | — | — | — | — | |
| EigenWDModel=Llama-3.1-8b2026.03 | — | 0.629 | — | — | — | — | — | |
| EigenWDModel=Qwen2-7b2026.03 | — | 0.668 | — | — | — | — | — | |
| EigenWDModel=Qwen2-72b2026.03 | — | 0.681 | — | — | — | — | — | |
| ERModel=Llama-3.2-3b2026.03 | — | — | 0.652 | — | — | — | — | |
| ERModel=Llama-2-7b2026.03 | — | — | 0.658 | — | — | — | — | |
| ERModel=Llama-3.1-8b2026.03 | — | — | 0.624 | — | — | — | — | |
| ERModel=Qwen2-7b2026.03 | — | — | 0.675 | — | — | — | — | |
| ERModel=Qwen2-72b2026.03 | — | — | 0.668 | — | — | — | — | |
| ESModel=Llama-3.2-3b2026.03 | — | — | — | 0.633 | — | — | — | |
| ESModel=Llama-2-7b2026.03 | — | — | — | 0.646 | — | — | — | |
| ESModel=Llama-3.1-8b2026.03 | — | — | — | 0.62 | — | — | — | |
| ESModel=Qwen2-7b2026.03 | — | — | — | 0.668 | — | — | — | |
| ESModel=Qwen2-72b2026.03 | — | — | — | 0.685 | — | — | — | |
| LNEModel=Llama-3.2-3b2026.03 | — | — | — | — | — | 0.629 | — | |
| LNEModel=Llama-2-7b2026.03 | — | — | — | — | — | 0.637 | — | |
| LNEModel=Llama-3.1-8b2026.03 | — | — | — | — | — | 0.548 | — | |
| LNEModel=Qwen2-7b2026.03 | — | — | — | — | — | 0.615 | — | |
| LNEModel=Qwen2-72b2026.03 | — | — | — | — | — | 0.606 | — | |
| LSModel=Llama-3.2-3b2026.03 | — | — | — | — | — | — | 0.656 | |
| LSModel=Llama-2-7b2026.03 | — | — | — | — | — | — | 0.634 | |
| LSModel=Llama-3.1-8b2026.03 | — | — | — | — | — | — | 0.605 | |
| LSModel=Qwen2-7b2026.03 | — | — | — | — | — | — | 0.645 | |
| LSModel=Qwen2-72b2026.03 | — | — | — | — | — | — | 0.672 |