AI Paper Atlas

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

Jailbreak Foundry: From Papers to Runnable Attacks for Reproducible Benchmarking

Zhicheng Fang, Jingjie Zheng, Chenxu Fu, Wei Xu

ICML 2026 · Adversarial robustness and security › Attacks on text and graphs · approach: LLM jailbreak and injection defense · 0 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
translate jailbreak papers into modules with the multi-agent JBF-FORGE pipelineJBF-FORGE · 30 attacks, matched settingsmean ASR deviation (pp): +0.26 mean deviation, range -16.0% to 20.0%Improved
provide the official runnable repository to the agent (with-repo vs paper-only) (ablation)JBF-FORGE · Five attacks (ABJ, EquaCode, GTA, SATA-MLM, SATA-ELP)mean ASR (%): 66.5% -> 86.3% (+19.8)Improved
add the official repo for GTA (ablation)JBF-FORGE · GTAASR (%): delta = +48.6%Improved
add the official repo for template-style attacks (EquaCode, ABJ) (ablation)JBF-FORGE · EquaCode, ABJASR (%): EquaCode +5.3%, ABJ +5.2%Improved
add enhanced refinement pass with Claude Code on hard subset (ablation)JBF-FORGE · Six attacks with largest gapsmean reproduced-reported gap (%): -16.2% -> -7.6%Improved
enhanced refinement pass on ISA (ablation)JBF-FORGE · ISAASR (%): 56.0 -> 56.0No clear effect
integrate attacks into JBF-LIB instead of original repos22 attacks with reference code · Lines of codetotal LOC: 22,714 -> 9,549 LOCImproved
integrate Mousetrap and TrojFill into JBF-LIBJBF-LIB · LOCLOCHurt
use victim-in-the-loop search instead of stateful selectionJailbreak attacks · AdvBench, all victimsmean ASR (%): 49.4% -> 60.3%Improved
use victim-in-the-loop instead of sampling searchJailbreak attacks on GPT-5.1 · AdvBench, GPT-5.1mean ASR (%): 50.9% -> 26.8%Hurt
use formal wrapper carriers instead of linguistic reframingJailbreak attacks · AdvBench, all victimsmean ASR (%): 39.3% -> 66.0%Improved

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