Object-Centric Representation Learning on COCO
40.8mBOiGLASS
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| GLASSCategory=Diffusion2026.03 | 40.8 | 48.7 | 39 | |
| SPOT+FS+RCCategory=Autoregressive Teacher-Student2026.03 | 35.7 | 45.3 | — | |
| SPOTCategory=Autoregressive Teacher-Student2026.03 | 35 | 44.7 | 33 | |
| Top-Down Slot Attention2026.03 | 33 | 40.3 | 31.2 | |
| Slot Merging2026.03 | 32.88 | 42.78 | 30.75 | |
| Slot MergingBackbone=ViT-B/16, Resolution=224x2242026.03 | 32.88 | 42.78 | — | |
| DIAS_image2026.03 | 32.8 | — | 30.1 | |
| DIASimage2026.03 | 32.8 | — | — | |
| DINOSAURCategory=DINOSAUR-Based2026.03 | 32.3 | 38.8 | — | |
| DINOSAUR2026.03 | 32.3 | 38.8 | — | |
| DINOSAUR+FS+RCCategory=DINOSAUR-Based2026.03 | 31.2 | 40.1 | — | |
| DINOSAUR+FS+RC2026.03 | 31.2 | 40.1 | — | |
| SlotDiffusionCategory=Diffusion2026.03 | 31 | 35 | — | |
| Stable-LSDCategory=Diffusion2026.03 | 30.4 | — | — | |
| MetaSlot-MLPCategory=Adaptive Slot Attention2026.03 | 29.5 | — | 27.9 | |
| MetaSlot-MLP2026.03 | 29.5 | — | — | |
| SLATE2026.03 | 29.1 | 33.6 | — | |
| SLATE2026.03 | 29.1 | 33.6 | — | |
| MetaSlot-TfdCategory=Adaptive Slot Attention2026.03 | 28.2 | — | 26.7 | |
| MetaSlot-Tfd2026.03 | 28.2 | — | — | |
| DINOSAUR-MLPCategory=DINOSAUR-Based2026.03 | 27.7 | 30.9 | — | |
| DINOSAUR-MLP2026.03 | 27.7 | 30.9 | — | |
| AdaSlotCategory=Adaptive Slot Attention2026.03 | 27.4 | — | — | |
| AdaSlot2026.03 | 27.4 | — | — | |
| MetaSlot-DfzCategory=Adaptive Slot Attention2026.03 | 27.2 | — | 25.8 | |
| SA2026.03 | 17.2 | 19.2 | — | |
| SA2026.03 | 17.2 | 19.2 | — |