List your company

Reliability Estimation on TruthfulQA

80.78AUROCDisAAD
47.926456.455764.98573.5143May 7, 2026

Evaluation results

MethodLinks
DisAADTarget Model=GPT-4, Setting=Black-box, O(1) Complexity=single-sample2026.0580.7890.79
DisAADTarget Model=LLaMA2-70B, Setting=Black-box, O(1) Complexity=single-sample2026.0580.1578.07
DisAADTarget Model=LLaMA2-70B, Setting=White-box, O(1) Complexity=single-sample2026.0580.1578.07
DisAADTarget Model=Qwen3-32B, Setting=White-box, O(1) Complexity=single-sample2026.0574.7883.1
LogTokUTarget Model=LLaMA2-70B, Setting=White-box, O(1) Complexity=single-sample2026.0572.1768.74
EntropyTarget Model=LLaMA2-70B, Setting=White-box, O(1) Complexity=single-sample2026.0567.3261.21
EigVTarget Model=LLaMA2-70B, Setting=Black-box, O(1) Complexity=multi-sample2026.0566.2177.03
EigVTarget Model=GPT-4, Setting=Black-box, O(1) Complexity=multi-sample2026.0566.0576.81
LogTokUTarget Model=Qwen3-32B, Setting=White-box, O(1) Complexity=single-sample2026.0566.0274.97
EntropyTarget Model=Qwen3-32B, Setting=White-box, O(1) Complexity=single-sample2026.0565.974.07
ProbabilityTarget Model=LLaMA2-70B, Setting=White-box, O(1) Complexity=single-sample2026.0565.8959.9
ProbabilityTarget Model=Qwen3-32B, Setting=White-box, O(1) Complexity=single-sample2026.0565.773.1
LESTarget Model=GPT-4, Setting=Black-box, O(1) Complexity=multi-sample2026.0565.177.2
LESTarget Model=LLaMA2-70B, Setting=Black-box, O(1) Complexity=multi-sample2026.0564.1976.39
SETarget Model=LLaMA2-70B, Setting=Black-box, O(1) Complexity=multi-sample2026.0553.9346.73
LENTarget Model=GPT-4, Setting=Black-box, O(1) Complexity=multi-sample2026.0553.5767.83
DSETarget Model=LLaMA2-70B, Setting=Black-box, O(1) Complexity=multi-sample2026.0552.5245.04
DSETarget Model=GPT-4, Setting=Black-box, O(1) Complexity=multi-sample2026.0551.2964.66
SETarget Model=GPT-4, Setting=Black-box, O(1) Complexity=multi-sample2026.0549.6263.54
LENTarget Model=LLaMA2-70B, Setting=Black-box, O(1) Complexity=multi-sample2026.0549.1946.52