Research map / Image generation / Text-to-image diffusion
Describe, Don't Dictate: Semantic Image Editing with Natural Language Intent
En Ci, Shanyan Guan, Yanhao Ge, Yilin Zhang, Wei Li, Zhenyu Zhang, Jian Yang, Ying Tai
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 description text instead of instruction text (ablation) | DescriptiveEdit fine-tuning · Emu Edit test | CLIP-T: 0.272 -> 0.284 | Improved |
| use description text instead of instruction text (ablation) | DescriptiveEdit fine-tuning · Emu Edit test | DINO-I: 0.739 -> 0.741 | Improved |
| use description text instead of instruction text (ablation) | DescriptiveEdit fine-tuning · Emu Edit test | SSIM: 0.551 -> 0.562 | Improved |
| use description text instead of instruction text (ablation) | DescriptiveEdit fine-tuning · Emu Edit test | PSNR: 18.123 -> 18.309 | Improved |
| replace direct replacement with learnable linear fusion (ablation) | Cross-Attentive UNet · Emu Edit test | PSNR: 13.78 -> 18.58 | Improved |
| replace direct addition with learnable linear fusion (ablation) | Cross-Attentive UNet · Emu Edit test | PSNR: 14.77 -> 18.58 | Improved |
| replace direct addition with learnable linear fusion (ablation) | Cross-Attentive UNet · Emu Edit test | SSIM: 0.4970 -> 0.6153 | Improved |
| replace direct addition with learnable linear fusion (ablation) | Cross-Attentive UNet · Emu Edit test | DINO-I: 0.7532 -> 0.7931 | Improved |
| replace direct addition with learnable linear fusion (ablation) | Cross-Attentive UNet · Emu Edit test | CLIP-T: 0.3052 -> 0.3162 | Improved |
| increase lambda_I (image guidance) (ablation) | DescriptiveEdit inference · seasonal editing (qualitative) | edit strength vs reference fidelity | Mixed |
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.
- AnyEdit — AnyEdit: Mastering Unified High-Quality Image Editing for Any Idea (CVPR 2025) · better on 4 of 4 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
- EditInfinity: Image Editing with Binary-Quantized Generative Models (NeurIPS 2025 · similarity 0.90)
- InsightEdit: Towards Better Instruction Following for Image Editing (CVPR 2025 · similarity 0.90)
- Imagine Before You Draw: Visual Prompt Engineering for Image Generation (NeurIPS 2026 · similarity 0.90)
- Textualize Visual Prompt for Image Editing via Diffusion Bridge (AAAI 2025 · similarity 0.90)
- FireEdit: Fine-grained Instruction-based Image Editing via Region-aware Vision Language Model (CVPR 2025 · similarity 0.89)
- PairEdit: Learning Semantic Variations for Exemplar-based Image Editing (NeurIPS 2025 · 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.
