List your company

Hallucination Detection on WikiBio Factual Sentence-level

44.23AUC-PRSelfCheckGPT
26.310830.962935.61540.2671May 6, 2026

Evaluation results

MethodLinks
SelfCheckGPT2026.0544.23
DS (llama-3)Backbone=Llama-32026.0542.4
DS (F2LLM)Backbone=F2LLM2026.0542
DS (avg)Backbone=average2026.0539.5
DS (Mistral)Backbone=Mistral2026.0538.1
DS (Qwen3)Backbone=Qwen32026.0537.6
DS (Jina-v5)Backbone=Jina-v52026.0537.2
Random2026.0527