Research map / Image generation / Editing and style transfer
Unpaired Visual Editing with Self-Consistent Flow Matching
Yoad Tewel, Yuval Atzmon, Gal Chechik, Lior Wolf
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 gradient routing (ablation) | Unpaired flow matching editor · GEdit-Bench | Source preservation: 7.183 -> 7.617 | Improved |
| add gradient routing (ablation) | Unpaired flow matching editor · GEdit-Bench | Edit success: 8.917 -> 8.317 | Hurt |
| add cycle loss (ablation) | Unpaired flow matching editor · GEdit-Bench | Source preservation: 7.233 -> 7.617 | Improved |
| add cycle loss (ablation) | Unpaired flow matching editor · GEdit-Bench | Edit success: 8.983 -> 8.317 | Hurt |
| add directional loss to the prior (ablation) | Unpaired flow matching editor · GEdit-Bench | Source preservation: 7.233 -> 7.617 | Improved |
| add bootstrapping (ablation) | Unpaired flow matching editor · GEdit-Bench | Edit success: 5.517 -> 8.317 | Improved |
| add bootstrapping (ablation) | Unpaired flow matching editor · GEdit-Bench | Source preservation: 7.050 -> 7.617 | Improved |
| add regularization loss (ablation) | Unpaired flow matching editor · GEdit-Bench | Edit success: 0.633 -> 8.317 | Improved |
| add random identity steps (ablation) | Unpaired flow matching editor · GEdit-Bench | Source preservation: 7.450 -> 7.617 | Improved |
| add random identity steps (ablation) | Unpaired flow matching editor · GEdit-Bench | Edit success: 8.413 -> 8.317 | No clear effect |
| use EMA for pseudo-target generation (ablation) | Video editing model · Video editing | DINO similarity: 0.656 -> 0.718 | Improved |
| use EMA for pseudo-target generation (ablation) | Video editing model · Video editing | CLIP directional similarity: 0.110 -> 0.104 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- FlowEdit — FlowEdit: Inversion-Free Text-Based Editing Using Pre-Trained Flow Models (ICCV 2025) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- DiTailed: Ensuring Visual Object Consistency in Text-Image-to-Image Flow Matching Models (ECCV 2026 · similarity 0.91)
- Learning an Image Editing Model without Image Editing Pairs (ICLR 2026 · similarity 0.90)
- Shifting the Breaking Point of Flow Matching for Multi-Instance Editing (ICML 2026 · similarity 0.89)
- SplitFlow: Flow Decomposition for Inversion-Free Text-to-Image Editing (NeurIPS 2025 · similarity 0.88)
- Latent Space Editing in Transformer-Based Flow Matching (AAAI 2024 · similarity 0.88)
- VINCIE-NExT: Unlocking Video Editing from Images via In-Context Modeling (NeurIPS 2026 · similarity 0.87)
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.
