AI Paper Atlas

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

New Wide-Net-Casting Jailbreak Attacks Risk Large Models

Qiuchi Xiang, Haoxuan Qu, Hossein Rahmani, Jun Liu

ICML 2026 · Adversarial robustness and security › Attacks on text and graphs · approach: LLM jailbreak and injection defense · 0 citations (OpenAlex)

Read the paper

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
wide-net-casting (OR over 4 LLMs) instead of single modelGCG · AdvBenchWASR vs ASR: 46.2% -> 75.0%Improved
wide-net-casting instead of single modelReMiss · AdvBenchWASR vs ASR: 86.5% -> 92.3%Improved
wide-net-casting over MLLMsMLAI+PixArt-alpha · MM-SafetyBenchWASR vs ASR: 77.9% -> 93.7%Improved
add SmoothLLM defenseReMiss group of 4 LLMs · AdvBenchWASR: 92.3% -> 61.5%Hurt
add RobustKV defenseReMiss group of 4 LLMs · AdvBenchWASR: 92.3% -> 56.1%Hurt
add VLGuard defenseMLAI+PixArt-alpha group · MM-SafetyBenchWASR: 93.7% -> 40.2%Hurt
restrict group to same LLaVA familyMLAI+PixArt-alpha · MM-SafetyBenchWASR: 93.7% -> 89.9%Hurt
restrict group to same family and versionMLAI+PixArt-alpha · MM-SafetyBenchWASR: 89.9% -> 41.1%Hurt
add joint specialized training (proposed)ReMiss baseline · AdvBench LLMsWASR: 92.3% -> 100%Improved
add joint specialized training (proposed)MLAI+PixArt-alpha · AdvBench MLLMs + VLGuardWASR: 37.5% -> 50.8%Improved
add theoretically derived update schedule over rule-based variants (ablation)Variant III · AdvBench MLLMsWASR: 96.7% -> 100%Improved
add linearly decreasing H_t instead of fixed H_t (ablation)Variant V · AdvBench MLLMsWASR: 98.0% -> 100%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