AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

VisualPrompter: Semantic-Aware Prompt Optimization with Visual Feedback for Text-to-Image Synthesis

Shiyu Wu, Mingzhen Sun, Weining Wang, Yequan Wang, Jing Liu

ICLR 2026 · Image generation › Text-to-image diffusion · approach: Personalized and region-controlled generation

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 changedWhereMeasuredEffect
add visual-feedback self-reflection and DSG-based concept expansionVisualPrompter · DSG-1k and TIFA, average over 4 generatorssemantic accuracy: 78.4 -> 83.0Improved
replace DSG-based concept expansion with direct LLM (Qwen) rewriting (ablation)VisualPrompter · DSG benchmarksemantic accuracy / aesthetic: 77.0 / 5.73 -> 73.0 / 5.45Hurt
direct Qwen prompt optimization without visual feedback (ablation)original prompts · DSG benchmarksemantic accuracy: 72.1 -> 67.8Hurt
add visual-feedback self-reflection to DSG prompt modification (ablation)VisualPrompter · DSG benchmarksemantic accuracy: 71.9 -> 77.0Improved
regenerate defective images without modifying prompts (ablation)original prompts · DSG benchmarksemantic accuracy: 72.1 -> 73.8Improved
replace the LLM decorator with NeuroPrompts (ablation)VisualPrompter · DSG benchmarkAesthetic Score: 5.81 -> 6.24Improved
replace the LLM decorator with NeuroPrompts (ablation)VisualPrompter · DSG benchmarksemantic consistencyHurt
apply prior prompt engineers (Promptist, BeautifulPrompt) to generators other than SD1.5Promptist, BeautifulPrompt · DSG benchmarkCLIP Score: Promptist Flux 31.58 -> 31.49Hurt
use VisualPrompter instead of original prompts on Flux-devFlux-dev · DSG benchmarkCLIP Score: 31.58 -> 33.81Improved
add VisualPrompter on top of generatorVisualPrompter · inference time, SD v1.5seconds: 7.0s (Promptist) vs 10.7sHurt

Later papers that compared with this method

1 later accepted paper reports VisualPrompter as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy