AI Paper Atlas

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

JANUS: A Lightweight Framework for Jailbreaking Text-to-Image Models via Distribution Optimization

Haolun Zheng, Yu He, Tailun Chen, Shuo Shao, Zhixuan Chu, Hongbin Zhou, Lan Tao, Zhan Qin, Kui Ren

CVPR 2026 · Image generation › Text-to-image diffusion · approach: T2I evaluation and controllable adapters · 2 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
use dual-distribution interference instead of a single distribution (ablation)JANUS · SD3.5LTASR-8 (%): 26.87% -> 44.50%Improved
use dual-distribution interference instead of a single distribution (ablation)JANUS · DALL-E3ASR-8 (%): 0.69% -> 1.39%Improved
use dual-distribution interference instead of a single distribution (ablation)JANUS · SD3.5LTTASR: 97.00% -> 94.25%Hurt
use dynamic harmfulness-based reward instead of fixed reward (ablation)JANUS · SD3.5LTNSFW score: 0.286 -> 0.329Improved
use dynamic harmfulness-based reward instead of fixed reward (ablation)JANUS · SD3.5LTASR-8 (%): 32.33% -> 44.50%Improved
use dynamic harmfulness-based reward instead of fixed reward (ablation)JANUS · DALL-E3NSFW score: 0.071 -> 0.082Improved
learn dynamic mixing alpha with RL instead of fixed alpha (ablation)JANUS · SD3.5LT and DALL-E3ASR and NSFW scoreImproved
use high fixed alpha (1.0) toward harmful distribution (ablation)JANUS · SD3.5LT and DALL-E3TASRHurt
use low fixed alpha toward clean distribution (ablation)JANUS · SD3.5LT and DALL-E3NSFW scoreHurt
use JANUS instead of PGJ on SD3.5LTJANUS · SD3.5LTASR-8 (%): 17.15 -> 43.15Improved
use JANUS instead of QFA on DALL-E3JANUS · DALL-E3ASR-8 (%): 1.53 -> 3.39Improved
use JANUS instead of QFA on SDXLJANUS · SDXLASR-8 (%): 24.26 -> 58.20Improved

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