Research map / Image generation / Text-to-image diffusion
FlowEdit: Inversion-Free Text-Based Editing Using Pre-Trained Flow Models
Vladimir Kulikov, Matan Kleiner, Inbar Huberman-Spiegelglas, Tomer Michaeli
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 |
|---|---|---|---|
| edit with FlowEdit direct ODE instead of editing by inversion | SD3 · Cat-to-dog | latent MSE: 2239 -> 1376 | Improved |
| edit with FlowEdit instead of inversion | SD3 · Cat-to-dog | LPIPS: 0.25 -> 0.15 | Improved |
| edit with FlowEdit instead of inversion | SD3 · Cat-to-dog | FID: 55.88 -> 51.14 | Improved |
| edit with FlowEdit instead of inversion | SD3 · Cat-to-dog | KID: 0.023 -> 0.017 | Improved |
| FlowEdit instead of ODE inversion on SD3 dataset | SD3 · Image editing dataset | LPIPS: 0.318 -> 0.181 | Improved |
| FlowEdit instead of ODE inversion on FLUX dataset | FLUX · Image editing dataset | LPIPS: 0.374 -> 0.223 | Improved |
| FlowEdit instead of ODE inversion on FLUX | FLUX · Image editing dataset | CLIP-T: 0.341 -> 0.337 | Hurt |
| provide a ChatGPT-varied source prompt (ablation) | FlowEdit · Source prompt ablation | DINO: 0.719 -> 0.713 | No clear effect |
| remove the source prompt (empty prompt) (ablation) | FlowEdit · Source prompt ablation | LPIPS: 0.181 -> 0.192 | No clear effect |
| increase n_avg from 1 toward 10 (ablation) | FlowEdit · SD3 | LPIPS | Improved |
| deviate from the exact step size dt in the update (ablation) | FlowEdit · editing | quality | Hurt |
Later papers that compared with this method
29 later accepted papers report FlowEdit as a baseline in their results tables; 28 of them report a better number on most of the results they share with it.
- Constrained Look-ahead Guidance for Interference-Aware Flow Editing (NeurIPS 2026 · better on 8 of 10 shared results)
- V-HOLD: Stabilizing Flow Trajectories to Rethink the Edit–Preservation Trade-off (ECCV 2026 · better on 7 of 9 shared results)
- Wavelet-Guided Semantic Signal Compensation for Inversion-Free Image Editing (ECCV 2026 · better on 8 of 9 shared results)
- FlowAlign: Trajectory-Regularized, Inversion-Free Flow-based Image Editing (ICLR 2026 · better on 7 of 8 shared results)
- FlowDC: Flow-Based Decoupling-Decay for Complex Image Editing (CVPR 2026 · better on 8 of 8 shared results)
- Training-Free Text-Guided Color Editing with Multi-Modal Diffusion Transformer (ICLR 2026 · better on 6 of 6 shared results)
- W-EDIT: A Wavelet-Based Frequency-Aware Framework for Text-Driven Image Editing (ICLR 2026 · better on 6 of 6 shared results)
- Delta Rectified Flow Sampling for Text-to-Image Editing (CVPR 2026 · better on 6 of 6 shared results)
21 more not listed.
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.
- SDEdit 0.2 — SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equations (ICLR 2022) · better on 10 of 10 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
- Inversion-Free Image Editing with Language-Guided Diffusion Models (CVPR 2024 · similarity 0.92)
- Semantic Image Inversion and Editing using Rectified Stochastic Differential Equations (ICLR 2025 · similarity 0.91)
- Lightning-Fast Image Inversion and Editing for Text-to-Image Diffusion Models (ICLR 2025 · similarity 0.90)
- Taming Rectified Flow for Inversion and Editing (ICML 2025 · similarity 0.90)
- SplitFlow: Flow Decomposition for Inversion-Free Text-to-Image Editing (NeurIPS 2025 · similarity 0.90)
- PnP Inversion: Boosting Diffusion-based Editing with 3 Lines of Code (ICLR 2024 · similarity 0.90)
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.
