List your company
ResearchFormerly WizWand

Hallucination Detection on SC-ChatGPT

74.9AUROCSELFCHECKGPT+f
49.73256.26662.869.334Jul 28, 2025
Updated 7mo ago

Evaluation results

MethodLinks
SELFCHECKGPT+fDetector=LLaMA 3.1, Future Context=true2025.0774.9
SELFCHECKGPTDetector=LLaMA 3.1, Future Context=false2025.0771.2
SELFCHECKGPT+fDetector=Gemma 3, Future Context=true2025.0771
SC+fDetector=LLaMA 3.1, Future Context=true2025.0769.2
DIRECT+fDetector=LLaMA 3.1, Future Context=true2025.0769
SELFCHECKGPTDetector=Gemma 3, Future Context=false2025.0768.8
DIRECTDetector=LLaMA 3.1, Future Context=false2025.0766.4
DIRECT+fDetector=Qwen 2.5, Future Context=true2025.0765.8
SC+fDetector=Gemma 3, Future Context=true2025.0765.6
SELFCHECKGPT+fDetector=Qwen 2.5, Future Context=true2025.0765.5
SELFCHECKGPTDetector=Qwen 2.5, Future Context=false2025.0764.4
DIRECTDetector=Qwen 2.5, Future Context=false2025.0763
SCDetector=LLaMA 3.1, Future Context=false2025.0762.5
SCDetector=Gemma 3, Future Context=false2025.0761.1
SC+fDetector=Qwen 2.5, Future Context=true2025.0759.5
DIRECT+fDetector=Gemma 3, Future Context=true2025.0758
DIRECTDetector=Gemma 3, Future Context=false2025.0752.5
SCDetector=Qwen 2.5, Future Context=false2025.0750.7