WeMMU: Enhanced Bridging of Vision-Language Models and Diffusion Models via Noisy Query Tokens
Open paper
Benchmark
Model
Rank
Results
image-editing-on-imgedit-data
WeMMU (Stage 3)
–
Overall: 3.31
Hybrid: 2.82
Action: 3.15
image-editing-on-imgedit-data
WeMMU (Stage 4)
–
Overall: 3.30
Hybrid: 2.78
Action: 3.17
Rank counts only results with a code link.