Research map / Image generation / Editing and style transfer
SuperEdit: Rectifying and Facilitating Supervision for Instruction-Based Image Editing
Ming Li, Xin Gu, Fan Chen, Xiaoying Xing, Longyin Wen, Chen Chen, Sijie Zhu
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 rectified editing instructions (40K) in place of original noisy instructions (ablation) | InstructPix2Pix-style SD1.5 · Real-Edit | following score: 2.45 -> 3.40 | Improved |
| add rectified editing instructions (ablation) | InstructPix2Pix-style SD1.5 · Real-Edit | preserving score: 3.27 -> 4.06 | Improved |
| add rectified editing instructions (ablation) | InstructPix2Pix-style SD1.5 · Real-Edit | quality score: 3.90 -> 4.01 | Improved |
| add contrastive instructions with triplet loss to rectified instructions (ablation) | SuperEdit · Real-Edit | following score: 3.40 -> 3.59 | Improved |
| add contrastive instructions with triplet loss (ablation) | SuperEdit · Real-Edit | preserving score: 4.06 -> 4.14 | Improved |
| add contrastive instructions with triplet loss (ablation) | SuperEdit · Real-Edit | quality score: 4.01 -> 4.01 | No clear effect |
| scale training data from 5k to 40k (ablation) | SuperEdit · Real-Edit | overall accuracy: 54.7% -> 69.7% | Improved |
| scale training data from 10k to 20k (ablation) | SuperEdit · Real-Edit | overall score: 3.63 -> 3.79 | Improved |
| use GPT-4o instead of open-source VLMs for rectification (ablation) | rectification pipeline · 100 samples | rectification success rate: 50.4% -> 76.2% | Improved |
| initialize from InstructDiffusion weights instead of SD1.5 (ablation) | SuperEdit · Real-Edit | preserving accuracy: 77% -> 83% | Improved |
| train at 256 resolution instead of 512 (ablation) | SuperEdit · Real-Edit | preserving accuracy: 77% -> 75% | Mixed |
| use SuperEdit (40K data, no extra module) instead of SmartEdit | SmartEdit · Real-Edit | overall score: 3.59 -> 3.91 | Improved |
Later papers that compared with this method
2 later accepted papers report SuperEdit* as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- LensStyle: Learning the Optical Aesthetics for Controllable Stylized Lens Effect Rendering (ECCV 2026 · better on 15 of 15 shared results)
- Semantically Aligned Gradient-Driven Context-Preserving Image Editing (ECCV 2026 · better on 6 of 7 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- InstructPix2Pix — InstructPix2Pix: Learning To Follow Image Editing Instructions (CVPR 2023) · better on 3 of 3 shared results
- HQ-Edit — HQ-Edit: A High-Quality Dataset for Instruction-based Image Editing (ICLR 2025) · better on 2 of 2 shared results
- HIVE — HIVE: Harnessing Human Feedback for Instructional Visual Editing (CVPR 2024) · 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
- InsightEdit: Towards Better Instruction Following for Image Editing (CVPR 2025 · similarity 0.91)
- ImgEdit: A Unified Image Editing Dataset and Benchmark (NeurIPS 2025 · similarity 0.90)
- AnyEdit: Mastering Unified High-Quality Image Editing for Any Idea (CVPR 2025 · similarity 0.90)
- FireEdit: Fine-grained Instruction-based Image Editing via Region-aware Vision Language Model (CVPR 2025 · similarity 0.90)
- Multi-Reward as Condition for Instruction-based Image Editing (ICLR 2025 · similarity 0.90)
- SmartEdit: Exploring Complex Instruction-based Image Editing with Multimodal Large Language Models (CVPR 2024 · similarity 0.89)
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.
