List your company

Hallucination Detection on True-False

89.7AUCPRSELFCHECKGPT
62.86869.83476.883.766Jul 28, 2025

Evaluation results

MethodLinks
SELFCHECKGPTDetector=Gemma 3, Future context=true2025.0789.7———————
SELFCHECKGPTDetector=LLaMA 3.1, Future context=true2025.0789.3———————
SELFCHECKGPTDetector=Gemma 3, Future context=false2025.0788.8———————
SELFCHECKGPTDetector=LLaMA 3.1, Future context=false2025.0788.1———————
DIRECTDetector=LLaMA 3.1, Future context=true2025.0787.1———————
SELFCHECKGPTDetector=Qwen 2.5, Future context=false2025.0786.5———————
SELFCHECKGPTDetector=Qwen 2.5, Future context=true2025.0785.4———————
DIRECTDetector=Gemma 3, Future context=true2025.0784.7———————
DIRECTDetector=LLaMA 3.1, Future context=false2025.0784.5———————
DIRECTDetector=Qwen 2.5, Future context=true2025.0783.9———————
DIRECTDetector=Gemma 3, Future context=false2025.0783.8———————
SCDetector=Gemma 3, Future context=false2025.0781.2———————
DIRECTDetector=Qwen 2.5, Future context=false2025.0780.5———————
SCDetector=Gemma 3, Future context=true2025.0780.4———————
SCDetector=LLaMA 3.1, Future context=true2025.0775.8———————
SCDetector=LLaMA 3.1, Future context=false2025.0771.4———————
SCDetector=Qwen 2.5, Future context=true2025.0764.7———————
SCDetector=Qwen 2.5, Future context=false2025.0763.9———————
EUBHDBackbone Model=LLaMA2-13B-Chat2024.11—56.8459.1958.3557.7655.8950.9656.5
LN-PPBackbone Model=LLaMA2-13B-Chat2024.11—55.3855.6453.975859.5456.1256.44
MINDBackbone Model=LLaMA2-13B-Chat2024.11—49.5747.0543.7250.1149.8645.0247.56
MMBackbone Model=LLaMA2-13B-Chat2024.11—53.353.6654.3853.2761.9650.8554.57
PRISM-MMBackbone Model=LLaMA2-13B-Chat2024.11—71.1188.3784.9870.9278.0479.9378.89
PRISM-SAPLMABackbone Model=LLaMA2-13B-Chat2024.11—74.0589.684.3569.9779.1882.0979.87
SAPLMABackbone Model=LLaMA2-13B-Chat2024.11—65.8470.6567.2164.7280.5462.5368.58