Research map / Image generation / Text-to-image diffusion
Adaptive Auxiliary Prompt Blending for Target-Faithful Diffusion Generation
Kwanyoung Lee, SeungJu Cha, Yebin Ahn, Hyunwoo Oh, Sungho Koh, Dong-Jin Kim
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 R2F heuristic prompt alternation with closed-form adaptive blending coefficient | SD3.0 rare concept generation · RareBench avg | GPT-4o T2I alignment: 75.7 -> 84.1 | Improved |
| use adaptive blending in FlowEdit editing | FlowEdit · FlowEdit dataset | structure preservation (DINO): 0.719 -> 0.814 | Improved |
| use adaptive blending in FlowEdit editing | FlowEdit · FlowEdit dataset | CLIP-T: 0.344 -> 0.341 | No clear effect |
| use adaptive coefficient instead of any fixed blending coefficient (ablation) | AAPB · RareBench avg | GPT-4o T2I alignment | Improved |
| use arbitrary random anchors instead of R2F's (ablation) | AAPB · RareBench avg | GPT-4o T2I alignment: 68.0 -> 71.0 (vs R2F with same anchors) | Improved |
| use GPT-4o generated anchors instead of human-annotated anchors (ablation) | AAPB · RareBench avg | GPT-4o T2I alignment: 82.6 -> 87.9 | Improved |
| use LLaMA3-8B anchors instead of human-annotated anchors (ablation) | AAPB · RareBench avg | GPT-4o T2I alignment: 82.6 -> 81.0 | No clear effect |
| set fixed coefficient to 0.0 (target only, no anchor) (ablation) | AAPB editing · FlowEdit dataset | CLIP-I vs CLIP-T trade-off | 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.
- FlowEdit — FlowEdit: Inversion-Free Text-Based Editing Using Pre-Trained Flow Models (ICCV 2025) · better on 4 of 5 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
- Textualize Visual Prompt for Image Editing via Diffusion Bridge (AAAI 2025 · similarity 0.91)
- Prompt-Free Diffusion: Taking "Text" out of Text-to-Image Diffusion Models (CVPR 2024 · similarity 0.90)
- DreamDistribution: Learning Prompt Distribution for Diverse In-distribution Generation (ICLR 2025 · similarity 0.90)
- Adapting Diffusion Models for Improved Prompt Compliance and Controllable Image Synthesis (NeurIPS 2024 · similarity 0.90)
- Noise Diffusion for Enhancing Semantic Faithfulness in Text-to-Image Synthesis (CVPR 2025 · similarity 0.90)
- Progressive Text-to-Image Diffusion with Soft Latent Direction (AAAI 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.
