Reliability Estimation on BioASQ
70.46AUROCDisAAD
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| DisAADTarget Model=LLaMA2-70B, Setting=Black-box, O(1) Complexity=single-sample2026.05 | 70.46 | 78.74 | |
| DisAADTarget Model=LLaMA2-70B, Setting=White-box, O(1) Complexity=single-sample2026.05 | 70.46 | 78.74 | |
| DisAADTarget Model=GPT-4, Setting=Black-box, O(1) Complexity=single-sample2026.05 | 69.93 | 87.42 | |
| SETarget Model=GPT-4, Setting=Black-box, O(1) Complexity=multi-sample2026.05 | 65.84 | 57.8 | |
| DisAADTarget Model=Qwen3-32B, Setting=White-box, O(1) Complexity=single-sample2026.05 | 65.71 | 84.8 | |
| EntropyTarget Model=Qwen3-32B, Setting=White-box, O(1) Complexity=single-sample2026.05 | 64.67 | 83.86 | |
| LogTokUTarget Model=Qwen3-32B, Setting=White-box, O(1) Complexity=single-sample2026.05 | 64.47 | 83.41 | |
| ProbabilityTarget Model=Qwen3-32B, Setting=White-box, O(1) Complexity=single-sample2026.05 | 63.93 | 83.33 | |
| EigVTarget Model=GPT-4, Setting=Black-box, O(1) Complexity=multi-sample2026.05 | 63.9 | 56.01 | |
| LENTarget Model=GPT-4, Setting=Black-box, O(1) Complexity=multi-sample2026.05 | 60.9 | 48.27 | |
| DSETarget Model=GPT-4, Setting=Black-box, O(1) Complexity=multi-sample2026.05 | 60.51 | 47.89 | |
| SETarget Model=LLaMA2-70B, Setting=Black-box, O(1) Complexity=multi-sample2026.05 | 59.94 | 48.25 | |
| DSETarget Model=LLaMA2-70B, Setting=Black-box, O(1) Complexity=multi-sample2026.05 | 59.83 | 44.43 | |
| EntropyTarget Model=LLaMA2-70B, Setting=White-box, O(1) Complexity=single-sample2026.05 | 59.67 | 69.24 | |
| LogTokUTarget Model=LLaMA2-70B, Setting=White-box, O(1) Complexity=single-sample2026.05 | 58.98 | 69.43 | |
| ProbabilityTarget Model=LLaMA2-70B, Setting=White-box, O(1) Complexity=single-sample2026.05 | 57.07 | 67.2 | |
| LENTarget Model=LLaMA2-70B, Setting=Black-box, O(1) Complexity=multi-sample2026.05 | 54.66 | 48.11 | |
| EigVTarget Model=LLaMA2-70B, Setting=Black-box, O(1) Complexity=multi-sample2026.05 | 53.15 | 46.22 | |
| LESTarget Model=GPT-4, Setting=Black-box, O(1) Complexity=multi-sample2026.05 | 53.15 | 46.05 | |
| LESTarget Model=LLaMA2-70B, Setting=Black-box, O(1) Complexity=multi-sample2026.05 | 52.85 | 45.73 |