Research map / Image generation / Fast diffusion sampling
SlerpFlow: Spherical Trajectory Correction for Rectified Flow Inversion
Wenbin Duan, Yan Shu, Zhuoyuan Fu, Fangmin Zhao, Yan Li, Yaru Zhao, Binyang Li
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 |
|---|---|---|---|
| replace Euler update with SlerpFlow spherical chordal update | FLUX.1-dev inversion · PIE-Bench reconstruction (unconditional) | PSNR: 11.10 -> 20.73 | Improved |
| replace FireFlow with SlerpFlow | FLUX.1-dev inversion · PIE-Bench reconstruction (conditional) | SSIM x100: 63.85 -> 79.35 | Improved |
| replace FireFlow with SlerpFlow | FLUX.1-dev inversion · PIE-Bench reconstruction (unconditional) | LPIPS x100: 31.09 -> 30.56 | Improved |
| replace FireFlow with SlerpFlow in editing | FLUX.1-dev editing · PIE-Bench editing | CLIP Similarity (Edited): 25.96 -> 26.38 | Improved |
| replace FireFlow with SlerpFlow in editing | FLUX.1-dev editing · PIE-Bench editing | Background PSNR: 22.47 -> 21.14 | Hurt |
| replace FireFlow with SlerpFlow in editing | FLUX.1-dev editing · PIE-Bench editing | Structure Distance x1000: 22.05 -> 23.35 | Hurt |
| use SlerpFlow at very few sampling steps | SlerpFlow · PIE-Bench reconstruction RMSE vs steps | reconstruction RMSE | Hurt |
| increase sampling steps for SlerpFlow (ablation) | SlerpFlow · PIE-Bench reconstruction RMSE vs steps | reconstruction RMSE | Improved |
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.
- FireFlow — FireFlow: Fast Inversion of Rectified Flow for Image Semantic Editing (ICML 2025) · better on 8 of 11 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
- Root-Selecting Fixed-Point Inversion for Rectified Flows via Trajectory Straightness (NeurIPS 2026 · similarity 0.88)
- Taming Rectified Flow for Inversion and Editing (ICML 2025 · similarity 0.88)
- FireFlow: Fast Inversion of Rectified Flow for Image Semantic Editing (ICML 2025 · similarity 0.87)
- Semantic Image Inversion and Editing using Rectified Stochastic Differential Equations (ICLR 2025 · similarity 0.85)
- Flow Priors for Linear Inverse Problems via Iterative Corrupted Trajectory Matching (NeurIPS 2024 · similarity 0.84)
- SCoT: Unifying Consistency Models and Rectified Flows via Straight-Consistent Trajectories (NeurIPS 2025 · similarity 0.84)
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.
