Research map / Image generation / Editing and style transfer
Language-Based Image Editing With Recurrent Attentive Models
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 recurrent attention to align text features with image regions (ablation) | CoSaL colorization model · CoSaL, 4 direct descriptions | average IoU: 0.2107 -> 0.4030 | Helped |
| Add recurrent attention to align text features with image regions (ablation) | CoSaL colorization model · CoSaL, 6 direct descriptions | average IoU: 0.2499 -> 0.5220 | Helped |
| Add recurrent attention to align text features with image regions (ablation) | CoSaL colorization model · CoSaL, 8 direct descriptions | average IoU: 0.3186 -> 0.7097 | Helped |
| Increase maximum inference steps from 1 to 4 with attention (ablation) | CoSaL colorization model · CoSaL, 4 direct descriptions | average IoU: 0.4030 -> 0.5033 | Helped |
| Increase maximum inference steps from 1 to 4 with attention (ablation) | CoSaL colorization model · CoSaL, 6 direct descriptions | average IoU: 0.5220 -> 0.5313 | Helped |
| Increase maximum inference steps from 1 to 4 with attention (ablation) | CoSaL colorization model · CoSaL, 8 direct descriptions | average IoU: 0.7097 -> 0.7017 | Hurt |
| Use recurrent attentive fusion rather than the comparison model | Language-based segmentation · ReferIt | overall IoU: 48.03% -> 50.09% | Helped |
| Use the caption during colorization instead of omitting text input | Oxford-102 image-to-image colorization · Oxford-102 Flowers test | caption consistency: 0.27 -> 0.849 | Helped |
| Use the caption during colorization instead of omitting text input | Oxford-102 image-to-image colorization · Oxford-102 Flowers test | image quality: 0.404 -> 0.598 | Helped |
Closest papers by meaning
- ImgEdit: A Unified Image Editing Dataset and Benchmark (NeurIPS 2025 · similarity 0.87)
- FOCUS: Unified Vision-Language Modeling for Interactive Editing Driven by Referential Segmentation (NeurIPS 2025 · similarity 0.86)
- Early Semantic Grounding in Image Editing Models for Zero-Shot Referring Image Segmentation (NeurIPS 2026 · similarity 0.86)
- PairEdit: Learning Semantic Variations for Exemplar-based Image Editing (NeurIPS 2025 · similarity 0.85)
- FireEdit: Fine-grained Instruction-based Image Editing via Region-aware Vision Language Model (CVPR 2025 · similarity 0.85)
- CAMILA: Context-Aware Masking for Image Editing with Language Alignment (NeurIPS 2025 · 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.