Hallucination Detection on BioASQ (test)
77.51AUROCQAODV⊥
Evaluation Results
| Method | Links | |
|---|---|---|
| QAODV⊥Model=Qwen3-30B-A3B, Zero-shot cross-domain=true2026.05 | 77.51 | |
| QAODV⊥Model=Qwen3-14B, Zero-shot cross-domain=true2026.05 | 76.73 | |
| QAODV⊥Model=Llama-2-7B-chat, Zero-shot cross-domain=true2026.05 | 75.63 | |
| QAODQ⊕V⊥Model=Qwen3-30B-A3B, Zero-shot cross-domain=true2026.05 | 74.27 | |
| QAODV⊥Model=gemma-2-2b, Zero-shot cross-domain=true2026.05 | 72.19 | |
| QAODQ⊕V⊥Model=Qwen3-14B, Zero-shot cross-domain=true2026.05 | 71.62 | |
| QAODQ⊕V⊥Model=gemma-2-2b, Zero-shot cross-domain=true2026.05 | 70.95 | |
| PikModel=Qwen3-30B-A3B, Zero-shot cross-domain=true2026.05 | 66.1 | |
| QAODQ⊕V⊥Model=Llama-2-7B-chat, Zero-shot cross-domain=true2026.05 | 65.28 | |
| SAPLMAModel=Qwen3-30B-A3B, Zero-shot cross-domain=true2026.05 | 63.63 | |
| MINDModel=Qwen3-30B-A3B, Zero-shot cross-domain=true2026.05 | 62.95 | |
| SAPLMAModel=gemma-2-2b, Zero-shot cross-domain=true2026.05 | 59.15 | |
| PikModel=Qwen3-14B, Zero-shot cross-domain=true2026.05 | 58.48 | |
| SAPLMAModel=Llama-2-7B-chat, Zero-shot cross-domain=true2026.05 | 56.76 | |
| PikModel=Llama-2-7B-chat, Zero-shot cross-domain=true2026.05 | 55.72 | |
| SAPLMAModel=Qwen3-14B, Zero-shot cross-domain=true2026.05 | 55.29 | |
| MINDModel=Qwen3-14B, Zero-shot cross-domain=true2026.05 | 51.99 | |
| MINDModel=gemma-2-2b, Zero-shot cross-domain=true2026.05 | 48.35 | |
| MINDModel=Llama-2-7B-chat, Zero-shot cross-domain=true2026.05 | 45.77 | |
| PikModel=gemma-2-2b, Zero-shot cross-domain=true2026.05 | 45.1 |