Visual Anomaly Segmentation on ClinicDB
86.6APAG-VAS
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| AG-VASBase Model=LLaVA-OneVision-7B2026.03 | 86.6 | 79.2 | 69.5 | |
| LISA*Base Model=LLaVA-OneVision-7B, Retrained=true2026.03 | 81 | 74.1 | 59.8 | |
| AG-VASBase Model=LLaVA-13B2026.03 | 68.3 | 65.5 | 50.1 | |
| LISA*Base Model=LLaVA-13B, Retrained=true2026.03 | 67.8 | 62.7 | 48.3 | |
| LISABase Model=LLaVA-13B2026.03 | 56.3 | 52.9 | 42.3 | |
| LISABase Model=LLaVA-OneVision-7B2026.03 | 56.1 | 54.4 | 32.6 | |
| Bayes-PFLBase Model=CLIP(ViT-L-14-336)2026.03 | 53.2 | 51.7 | 21 | |
| LISABase Model=LLaVA-7B2026.03 | 50 | 45.3 | 36.4 | |
| Bayes-PFL*Base Model=CLIP(ViT-L-14-336), Retrained=true2026.03 | 47.6 | 50.7 | 34.4 | |
| AnomalyCLIP*Base Model=CLIP(ViT-L-14-336), Retrained=true2026.03 | 44.1 | 47.3 | 29.4 | |
| AnomalyCLIPBase Model=CLIP(ViT-L-14-336)2026.03 | 42.2 | 45.9 | 30 | |
| PaDTBase Model=Qwen2.5-VL-7B2026.03 | 39.8 | 51.4 | 49.8 | |
| APRIL-GANBase Model=CLIP(ViT-L-14-336)2026.03 | 38.8 | 44.4 | 25.3 | |
| WinCLIPBase Model=CLIP(ViT-B-16-240)2026.03 | 19.4 | 27.3 | 0 | |
| PixelLMBase Model=LLaVA-13B2026.03 | 13 | 20.9 | 14.6 |