Multi-image hallucination evaluation on MUIRBench
62Accuracy
Qwen2.5-VL + CAPL
Updated 1mo ago
Evaluation Results
| Method | Links | |
|---|---|---|
2026.03 | 62 | |
2026.03 | 60.57 | |
2026.03 | 58.42 | |
2026.03 | 57.84 | |
2026.03 | 52.12 | |
2026.03 | 48.54 | |
2026.03 | 45.61 | |
2026.03 | 39.57 | |
2026.03 | 30.96 | |
2026.03 | 30.85 | |
2026.03 | 30.5 | |
2026.03 | 29.85 |