AI Paper Atlas

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

Value-Aligned Prompt Moderation via Zero-Shot Agentic Rewriting for Safe Image Generation

Xin Zhao, Xiaojun Chen, Bingshan Liu, Zeyao Liu, Zhendong Zhao, Xiaoyan Gu

AAAI 2026 · Image generation › Text-to-image diffusion · approach: Layout-to-image control · 1 citations (OpenAlex)

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 VALOR prompt rewriting in front of the T2I modelSDV1.4 · I2P-SexualSAFE rate: 67.9% -> 92.7%Improved
add VALOR prompt rewriting in front of the T2I modelSDV1.4 · 4chanSAFE rate: 83.2% -> 99.6%Improved
add word-level keyword detection (ablation)VALOR · Value-412ACC: 88.8% -> 90.5%Improved
add semantic-level detection (ablation)VALOR · Value-412FNR: 16.6% -> 11.2%Improved
add value-level detection (ablation)VALOR · Value-412ACC: 88.8% -> 90.5%Improved
add detection gating before rewriting instead of rewriting everything (ablation)VALOR · Value-412FPR: 100.0% -> 6.5%Improved
use a small rewriter (Qwen1.5-1.8B-Chat) instead of larger LLMsVALOR · Intent-452FNR: up to 33.3%Hurt
add VALOR rewriting on benign promptsSDV1.4 · COCO (benign)FID: 17.04 -> 15.89Improved
add LLM rewriting step to the pipelineVALOR · Average time per promptlatency overhead: 0.575 s to 2.54 s per promptHurt
use VALOR instead of baseline defenses on jailbreak datasetsSDV1.4 · Jailbreak datasets (SneakyPrompt, QF-Attack, MMP-Attack, MMA-Diffusion)NRR: 98.0% to 100.0% versus lowest baseline 12.7%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.

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