Hallucination detection on llava
96.5AUC ROC
HALLUSHIFT++
Updated 1mo ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2025.12 | 96.5 | 94.8 | 88.6 | |
2025.12 | 88.2 | 76.4 | 52.2 | |
2025.12 | 85.9 | 68.3 | 63.4 | |
2025.12 | 84.8 | 67.6 | 61.5 | |
2025.12 | 84.3 | 64.8 | 58.3 | |
2025.12 | 84 | 66.8 | 59.6 | |
2025.12 | 83.8 | 70.6 | 61.8 | |
2025.12 | 82.8 | 67.9 | 59.6 | |
2025.12 | 71.1 | 53.2 | 61.6 | |
2025.12 | 61.7 | 57.2 | 53 | |
2025.12 | 58.8 | 74 | 79.5 | |
2025.12 | 58.4 | 47.5 | 49.5 | |
2025.12 | 57.9 | 54.3 | 50.2 | |
2025.12 | 55.5 | 51.7 | 44.8 | |
2025.12 | 55.4 | 43.4 | 42.3 | |
2025.12 | 51.8 | 40.2 | 39.5 | |
2025.12 | 51.6 | 34.3 | 43.3 | |
2025.12 | 51.5 | 25 | 33.3 | |
2025.12 | 48.2 | 31.5 | 40.1 |