Image Generation on ImageNet-1K 10K subset 256x256
3.43FIDOurs
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| OursData Budget=0.8% (10K), Backbone=DiT-L/2, Training Iterations=100K, Resolution=256x256, Evaluation Samples=50K2026.06 | 3.43 | 414.3 | 78 | 28 | |
| D2CData Budget=0.8% (10K), Backbone=DiT-L/2, Training Iterations=100K, Resolution=256x256, Evaluation Samples=50K2026.06 | 4.2 | 283.6 | 72 | 24 | |
| DQData Budget=0.8% (10K), Backbone=DiT-L/2, Training Iterations=100K, Resolution=256x256, Evaluation Samples=50K2026.06 | 4.56 | 267.8 | 72 | 25 | |
| Random†Data Budget=0.8% (10K), Backbone=DiT-L/2, Training Iterations=100K, Resolution=256x256, Evaluation Samples=50K2026.06 | 4.63 | 263.1 | 70 | 26 | |
| CCSData Budget=0.8% (10K), Backbone=DiT-L/2, Training Iterations=100K, Resolution=256x256, Evaluation Samples=50K2026.06 | 5.45 | 364.9 | 77 | 21 |