AI Paper Atlas

Research map / Adversarial robustness and security / Attacks on text and graphs

JailbreakDiffBench: A Comprehensive Benchmark for Jailbreaking Diffusion Models

Xiaolong Jin, Zixuan Weng, Hanxi Guo, Chenlong Yin, Siyuan Cheng, Guangyu Shen, Xiangyu Zhang

ICCV 2025 · Adversarial robustness and security › Attacks on text and graphs · approach: Diffusion-based adversarial purification · 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
add in-context examples to GPT-4o-mini prompt checker (ablation)GPT-4o-mini · JDBench promptsaccuracy: 0.8540 -> 0.8720Improved
add in-context examples to GPT-4o prompt checker (ablation)GPT-4o · JDBench promptsFPR: 0.5481 -> 0.2869Improved
add in-context examples to GPT-4o prompt checker (ablation)GPT-4o · JDBench promptsaccuracy: 0.6184 -> 0.7691Improved
add in-context examples to GPT-4o-mini (ablation)GPT-4o-mini · Copyright categoryrecall: 0.562 -> 0.342Hurt
use LLM checker GPT-4o instead of Multihead Detector for imagesMultihead Detector · image checkingaccuracy: 0.7571 -> 0.9203Improved
use rule-based keyword matcher vs BERT classifierNSFW-Word-Matcher · JDBench promptsF1: 0.2635 -> 0.7991Hurt
use larger GPT-4o instead of GPT-4o-mini as prompt checkerGPT-4o-mini · JDBench promptsFPR: 0.1376 -> 0.5481Hurt
black-box PGJ attack vs white-box RingABell against checkersRingABell · SD-v1.5BPR: 0.50% -> 51.75%Improved
bypassing checkers does not translate to harmful imagesPGJ · SD-v1.5ASR: 51.75% -> 5.75%Hurt
newer SD-3.5-Large-Turbo vs SD-v1.5 vulnerabilitySD-v1.5 · DACA attackBPR: 57.00% -> 56.25%No clear effect

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