11 papers
From Latents to Wires: Surgical Post-Editing on Large Language Models
Jiankai Jin, Xiangzheng Zhang, Zhao Liu +4
Given a large language model (LLM), can whoever holds the weights name a semantic target (e.g., the model's identity), locate the model components that produce it, and edit them so…
ReOrder-OPD:Reliability-Aware Prompt Ordering for On-Policy Distillation
Ximo Zhu, Ruiqi Liu, Rong Wang +8
On-policy distillation (OPD) applies token-level teacher supervision to student-generated trajectories, but this supervision is not always reliable. Existing methods use local conf…
Dynamic Defense Profiling Enables Cognitive Jailbreak of Text-to-Image Models
Dongdong Yang, Deyue Zhang, Zhao Liu +5
Text-to-Image (T2I) generative models have achieved remarkable progress in synthesizing high-quality visual content, yet they remain vulnerable to adversarial misuse, particularly…
ElephantAgent: Contextual State Continuity in Agentic Systems
Jiankai Jin, Xiangzheng Zhang, Zhao Liu +4
Agentic systems enhance their capabilities by invoking external tools and maintaining persistent memory. However, these external dependencies introduce novel attack surfaces. Recen…
DMN: A Compositional Framework for Jailbreaking Multimodal LLMs with Multi-Image Inputs
Wenzhuo Xu, Zhipeng Wei, Zonghao Ying +4
Multimodal Large Language Models (MLLMs) are vulnerable to jailbreak attacks, which can elicit harmful responses from MLLMs. Many MLLMs support multi-image inputs, inadvertently in…
Reasoning-Oriented Programming: Chaining Semantic Gadgets to Jailbreak Large Vision Language Models
Quanchen Zou, Moyang Chen, Zonghao Ying +6
Large Vision-Language Models (LVLMs) undergo safety alignment to suppress harmful content. However, current defenses predominantly target explicit malicious patterns in the input r…
Two Frames Matter: A Temporal Attack for Text-to-Video Model Jailbreaking
Moyang Chen, Zonghao Ying, Wenzhuo Xu +4
Recent text-to-video (T2V) models can synthesize complex videos from lightweight natural language prompts, raising urgent concerns about safety alignment in the event of misuse in…
Robust Privacy: Inference-Stage Privacy through Certified Robustness
Jiankai Jin, Xiangzheng Zhang, Zhao Liu +4
An adversary observing a model's released prediction can infer sensitive attributes of the queried input, or even reconstruct representatives of the model's training data. The infe…
Sequential Comics for Jailbreaking Multimodal Large Language Models via Structured Visual Storytelling
Deyue Zhang, Dongdong Yang, Junjie Mu +6
Multimodal large language models (MLLMs) exhibit remarkable capabilities but remain susceptible to jailbreak attacks exploiting cross-modal vulnerabilities. In this work, we introd…
PRISM: Programmatic Reasoning with Image Sequence Manipulation for LVLM Jailbreaking
Quanchen Zou, Zonghao Ying, Moyang Chen +7
The increasing sophistication of large vision-language models (LVLMs) has been accompanied by advances in safety alignment mechanisms designed to prevent harmful content generation…
Probabilistic Modeling of Jailbreak on Multimodal LLMs: From Quantification to Application
Wenzhuo Xu, Zhipeng Wei, Xiongtao Sun +5
Recently, Multimodal Large Language Models (MLLMs) have demonstrated their superior ability in understanding multimodal content. However, they remain vulnerable to jailbreak attack…