Hallucination Detection on SelfCheckGPT
67.3AUROCSELFCHECKGPT+f
Evaluation Results
| Method | Links | |
|---|---|---|
| SELFCHECKGPT+fDetector=LLaMA 3.1, Future Context=true2025.07 | 67.3 | |
| SELFCHECKGPTDetector=LLaMA 3.1, Future Context=false2025.07 | 67 | |
| DIRECT+fDetector=LLaMA 3.1, Future Context=true2025.07 | 65.3 | |
| DIRECT+fDetector=Qwen 2.5, Future Context=true2025.07 | 62.9 | |
| DIRECTDetector=LLaMA 3.1, Future Context=false2025.07 | 62.1 | |
| DIRECTDetector=Qwen 2.5, Future Context=false2025.07 | 60.6 | |
| SELFCHECKGPT+fDetector=Gemma 3, Future Context=true2025.07 | 58.3 | |
| SC+fDetector=LLaMA 3.1, Future Context=true2025.07 | 57.7 | |
| DIRECT+fDetector=Gemma 3, Future Context=true2025.07 | 56.6 | |
| SC+fDetector=Gemma 3, Future Context=true2025.07 | 55.2 | |
| SC+fDetector=Qwen 2.5, Future Context=true2025.07 | 54.2 | |
| DIRECTDetector=Gemma 3, Future Context=false2025.07 | 53.7 | |
| SCDetector=LLaMA 3.1, Future Context=false2025.07 | 53.2 | |
| SCDetector=Gemma 3, Future Context=false2025.07 | 52.9 | |
| SELFCHECKGPTDetector=Qwen 2.5, Future Context=false2025.07 | 51.7 | |
| SELFCHECKGPT+fDetector=Qwen 2.5, Future Context=true2025.07 | 51.7 | |
| SCDetector=Qwen 2.5, Future Context=false2025.07 | 50.5 | |
| SELFCHECKGPTDetector=Gemma 3, Future Context=false2025.07 | 49.3 |