DreamBench++
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Personalized Text-to-Image Generation | DreamBench++ Single-subject | CP0.738 | 18 | |
| Multimodal Image Generation | DreamBench++ (test) | CP (Animal)95 | 14 | |
| Image-conditioned semantic transport | DreamBench++ (test) | CLIP T*I Score66.8 | 14 | |
| Personalized Image Generation | DreamBench++ GPT-4o score evaluation (test) | CP (Animal)90 | 8 | |
| Personalized Text-to-Image Generation | DreamBench++ (test) | CP Score3.733 | 8 | |
| Subject-Driven Text-to-Image Generation | DreamBench++ | CLIP-B-I0.801 | 6 | |
| Subject-driven generation | DreamBench++ 1.0 (test) | GPT-4o Score3.119 | 4 | |
| Zero-shot concept preservation | DreamBench++ | DINO-I50.9 | 4 |
Showing 8 of 8 rows