Research map / Image generation / Text-to-image diffusion
MaskDiffusion: Boosting Text-to-Image Consistency with Conditional Mask
Yupeng Zhou, Daquan Zhou, Yaxing Wang, Jiashi Feng, Qibin Hou
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add conditional cross-attention mask | Stable Diffusion · User study (simple prompts) | supporting rate: 3.10% -> 83.16% | Improved |
| add conditional cross-attention mask | Stable Diffusion · User study (complex prompts) | supporting rate: 9.16% -> 74.77% | Improved |
| add conditional cross-attention mask | Stable Diffusion · CLIP score | CLIP score: 26.16 -> 26.82 | Improved |
| add conditional cross-attention mask | Stable Diffusion · Inference time | seconds: 81s -> 81s | No clear effect |
| use mask instead of Attend-and-Excite iterative optimization | Attend-and-Excite · Inference time | seconds: 215s -> 81s | Improved |
| use MaskDiffusion instead of Composable Diffusion | Composable Diffusion · CLIP score | CLIP score: 26.59 -> 26.82 | Improved |
| add momentum updating of cross-attention maps (ablation) | MaskDiffusion · Qualitative object presence | qualitative | Improved |
| add mask on 16x16 cross-attention resolution (ablation) | MaskDiffusion · Qualitative consistency | qualitative | Improved |
Closest papers by meaning
- CONFORM: Contrast is All You Need for High-Fidelity Text-to-Image Diffusion Models (CVPR 2024 · similarity 0.89)
- Compositional Text-to-Image Synthesis with Attention Map Control of Diffusion Models (AAAI 2024 · similarity 0.89)
- Noise Diffusion for Enhancing Semantic Faithfulness in Text-to-Image Synthesis (CVPR 2025 · similarity 0.88)
- Seg4Diff: Unveiling Open-Vocabulary Semantic Segmentation in Text-to-Image Diffusion Transformers (NeurIPS 2025 · similarity 0.88)
- $\text{I}^2\text{AM}$: Interpreting Image-to-Image Latent Diffusion Models via Bi-Attribution Maps (ICLR 2025 · similarity 0.88)
- I Think, Therefore I Diffuse: Enabling Multimodal In-Context Reasoning in Diffusion Models (ICML 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
