Object Detection on S-DGOD
39.1mAP (Average)VLTDet
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| VLTDetEncoder=ViT-L-14, Init=EVA02-CLIP2023.12 | 39.1 | 56.6 | 44.4 | 43.6 | 26.6 | 41.8 | — | |
| PODAEncoder=R-101, Init=CLIP2023.12 | 37.1 | — | 43.4 | 40.2 | 20.5 | 44.4 | — | |
| VLTDetEncoder=R-101, Init=CLIP2023.12 | 36.9 | 60.5 | 44.6 | 38.4 | 22.1 | 42.3 | — | |
| Fine-tuning2026.03 | 36.37 | — | — | — | — | — | 57.83 | |
| G-NASEncoder=R-101, Init=IN1k2023.12 | 33.5 | 58.4 | 45 | 35.1 | 17.4 | 36.4 | — | |
| PDDOCEncoder=R-101, Init=IN1k2023.12 | 32.6 | 53.6 | 38.5 | 33.7 | 19.2 | 39.1 | — | |
| ABRA2026.03 | 32.35 | — | — | — | — | — | 53.81 | |
| Source2026.03 | 31.75 | — | — | — | — | — | 51.41 | |
| CLIP The GAPEncoder=R-101, Init=CLIP2023.12 | 31.6 | 51.3 | 36.9 | 32.3 | 18.7 | 38.5 | — | |
| S-DGODEncoder=R-101, Init=IN1k2023.12 | 28.7 | 56.1 | 36.6 | 28.2 | 16.6 | 33.5 | — | |
| Task Analogy2026.03 | 26.49 | — | — | — | — | — | 41.27 | |
| Zero shot2026.03 | 26.36 | — | — | — | — | — | 41.1 | |
| ParamΔ2026.03 | 17.68 | — | — | — | — | — | 26.68 |