List your company

Hallucination Detection on SelfAware

0.587AUROCNon-Contradiction
0.478840.506920.5350.56308Jun 1, 2026

Evaluation results

MethodLinks
Non-ContradictionModel=Llama-3-8B, Scorer Family=BB2026.060.587
GPT-4oModel=Llama-3-8B, Scorer Family=Judge2026.060.578
SonnetModel=GPT-4o, Scorer Family=Judge2026.060.528
SonnetModel=Gemini-2.5, Scorer Family=Judge2026.060.528
P(True)Model=Gemini-2.5, Scorer Family=WB2026.060.51
Sequence ProbabilityModel=Llama-3-8B, Scorer Family=WB2026.060.502
Exact MatchModel=GPT-4o, Scorer Family=BB2026.060.499
Exact MatchModel=Gemini-2.5, Scorer Family=BB2026.060.499
Probability MarginModel=GPT-4o, Scorer Family=WB2026.060.483