Research map / Image generation / Editing and style transfer
Editing Everything Everywhere All at Once
Fabio Quattrini, Carmine Zaccagnino, Enis Simsar, Marta Gazulla, Rita Cucchiara, Alessio Tonioni, Silvia Cascianelli
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 |
|---|---|---|---|
| use strict segmentation masks instead of bounding boxes (ablation) | MICE (FLUX.2 klein 4B) · MICE-Bench | MAEB: 10.66 -> 8.15 | Improved |
| use Gaussian smoothing on masks (ablation) | MICE (FLUX.2 klein 4B) · MICE-Bench | Attempt rate (%): 95.36 -> 87.30 | Hurt |
| add instance-aware smoothing over naive Gaussian smoothing (ablation) | MICE (FLUX.2 klein 4B) · MICE-Bench | Localized CLIP score: 26.19 -> 26.85 | Improved |
| add instance-aware smoothing over naive Gaussian smoothing (ablation) | MICE (FLUX.2 klein 4B) · MICE-Bench | Attempt rate (%): 87.30 -> 92.48 | Improved |
| add instance-aware smoothing over original masks (ablation) | MICE (FLUX.2 klein 4B) · MICE-Bench | Localized CLIP score: 26.63 -> 26.85 | Improved |
| add MICE to FLUX.2 klein 4B | FLUX.2 [klein] 4B · MICE-Bench | Localized CLIP score: 25.97 -> 26.85 | Improved |
| add MICE to FLUX.2 klein 9B | FLUX.2 [klein] 9B · MICE-Bench | Localized CLIP score: 26.34 -> 27.79 | Improved |
| add MICE to FLUX.1 Kontext | FLUX.1 Kontext · MICE-Bench | MAEB: 16.75 -> 6.61 | Improved |
| add MICE to FLUX.2-Dev | FLUX.2-Dev · LoMOE-Bench | Target CLIP score: 27.03 -> 26.44 | Hurt |
| add MICE to FLUX.2-Dev | FLUX.2-Dev · LoMOE-Bench | MAEB: 10.48 -> 11.50 | Hurt |
| use MICE biasing instead of IDAttn on same backbone | FLUX.2 [klein] 4B · MICE-Bench | Attempt rate (%): 77.05 -> 92.48 | Improved |
| use SAM3 predicted masks instead of ground-truth masks | MICE-9B · LoMOE-Bench | Localized CLIP score: 30.30 -> 30.41 | Improved |
Closest papers by meaning
- Improving Editability in Image Generation with Layer-wise Memory (CVPR 2025 · similarity 0.88)
- ParallelEdits: Efficient Multi-Aspect Text-Driven Image Editing with Attention Grouping (NeurIPS 2024 · similarity 0.87)
- UltraEdit: Instruction-based Fine-Grained Image Editing at Scale (NeurIPS 2024 · similarity 0.86)
- FireEdit: Fine-grained Instruction-based Image Editing via Region-aware Vision Language Model (CVPR 2025 · similarity 0.86)
- Edit-R2: Context-Aware Reinforcement Learning for Multi-Turn Image Editing (NeurIPS 2026 · similarity 0.86)
- Balancing Preservation and Modification: A Region and Semantic Aware Metric for Instruction-Based Image Editing (ICML 2025 · similarity 0.86)
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.
