Research map / Image generation / Text-to-image diffusion
LOCATEdit: Graph Laplacian Optimized Cross Attention for Localized Text-Guided Image Editing
Achint Soni, Meet Soni, Sirisha Rambhatla
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 diagonal weighting matrix (ablation) | LOCATEdit (PnP-I, DDIM(50)) · PIE-Bench | Structure Distance x10^3 (lower better): 8.68 -> 13.19 | Hurt |
| add diagonal weighting matrix (ablation) | LOCATEdit (PnP-I, DDIM(50)) · PIE-Bench | CLIP Similarity (Edited): 22.34 -> 23.02 | Improved |
| add diagonal weighting matrix (ablation) | LOCATEdit (PnP-I, DDIM(50)) · PIE-Bench | PSNR: 29.59 -> 29.20 | Hurt |
| add symmetric self-attention (ablation) | LOCATEdit (PnP-I, DDIM(50)) · PIE-Bench | Structure Distance x10^3 (lower better): 8.59 -> 13.19 | Hurt |
| add symmetric self-attention (ablation) | LOCATEdit (PnP-I, DDIM(50)) · PIE-Bench | SSIM: 85.66 -> 86.53 | Improved |
| add symmetric self-attention (ablation) | LOCATEdit (PnP-I, DDIM(50)) · PIE-Bench | CLIP Similarity (Edited): 22.26 -> 23.02 | Improved |
| add alpha-based control (ablation) | LOCATEdit (PnP-I, DDIM(50)) · PIE-Bench | CLIP Similarity (Whole): 25.31 -> 25.96 | Improved |
| add alpha-based control (ablation) | LOCATEdit (PnP-I, DDIM(50)) · PIE-Bench | Structure Distance x10^3 (lower better): 8.86 -> 13.19 | Hurt |
| use high alpha instead of default (ablation) | LOCATEdit (PnP-I, DDIM(50)) · PIE-Bench | CLIP Similarity (Edited): 23.02 -> 23.22 | Improved |
| use high alpha instead of default (ablation) | LOCATEdit (PnP-I, DDIM(50)) · PIE-Bench | PSNR: 29.20 -> 24.27 | Hurt |
| use high alpha instead of default (ablation) | LOCATEdit (PnP-I, DDIM(50)) · PIE-Bench | LPIPS x10^3 (lower better): 41.60 -> 87.37 | Hurt |
| replace ViMAEdit with LOCATEdit | PnP-I, DDIM(50) pipeline · PIE-Bench | LPIPS x10^3 (lower better): 43.07 -> 41.60 | 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.
- LEDITS++ — LEDITS++: Limitless Image Editing using Text-to-Image Models (CVPR 2024) · better on 3 of 3 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
- Towards Understanding Cross and Self-Attention in Stable Diffusion for Text-Guided Image Editing (CVPR 2024 · similarity 0.89)
- BARET: Balanced Attention Based Real Image Editing Driven by Target-Text Inversion (AAAI 2024 · similarity 0.87)
- Towards Efficient Diffusion-Based Image Editing with Instant Attention Masks (AAAI 2024 · similarity 0.87)
- On Mechanistic Knowledge Localization in Text-to-Image Generative Models (ICML 2024 · similarity 0.86)
- FireEdit: Fine-grained Instruction-based Image Editing via Region-aware Vision Language Model (CVPR 2025 · similarity 0.86)
- AdapEdit: Spatio-Temporal Guided Adaptive Editing Algorithm for Text-Based Continuity-Sensitive Image Editing (AAAI 2024 · similarity 0.85)
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.
