Chaowei Xiao

Doxing via the Lens: Revealing Location-related Privacy Leakage on Multi-modal Large Reasoning Models

ARMOR: Aligning Secure and Safe Large Language Models via Meticulous Reasoning

A2ASecBench: A Protocol-Aware Security Benchmark for Agent-to-Agent Multi-Agent Systems

Concept Concentration for Faithful Representation Intervention

TrustGen: A Platform of Dynamic Benchmarking on the Trustworthiness of Generative Foundation Models

Computer-Use Agent Frameworks Can Expose Realistic Risks Through Tactics, Techniques, and Procedures

PRISM: Robust VLM Alignment with Principled Reasoning for Integrated Safety in Multimodality

Mitigating Indirect Prompt Injection via Instruction-Following Intent Analysis

ReasAlign: Reasoning Enhanced Safety Alignment against Prompt Injection Attack

VLA-Risk: Benchmarking Vision-Language-Action Models with Physical Robustness

SAFEVISION: Efficient Image Guardrail with Robust Policy Adherence and Explainability

AutoDAN-Turbo: A Lifelong Agent for Strategy Self-Exploration to Jailbreak LLMs

T-Stitch: Accelerating Sampling in Pre-Trained Diffusion Models with Trajectory Stitching

Robust Representation Consistency Model via Contrastive Denoising

EIA: ENVIRONMENTAL INJECTION ATTACK ON GENERALIST WEB AGENTS FOR PRIVACY LEAKAGE

DRIFT: Dynamic Rule-Based Defense with Injection Isolation for Securing LLM Agents

JailDAM: Jailbreak Detection with Adaptive Memory for Vision-Language Model

Sample-specific Noise Injection for Diffusion-based Adversarial Purification

DataGen: Unified Synthetic Dataset Generation via Large Language Models

Can Watermarks be Used to Detect LLM IP Infringement For Free?

LeanAgent: Lifelong Learning for Formal Theorem Proving

Sample-specific Noise Injection for Diffusion-based Adversarial Purification

SafeVision: Efficient Image Guardrail with Robust Policy Adherence and Explainability

Benchmarking Vision Language Model Unlearning via Fictitious Facial Identity Dataset

Prompt Injection Benchmark for Foundation Model Integrated Systems

MuirBench: A Comprehensive Benchmark for Robust Multi-image Understanding

MetaAgent: Automatically Constructing Multi-Agent Systems Based on Finite State Machines

Can Editing LLMs Inject Harm?

MetaAgent: Automatically Building Multi-Agent System based on Finite State Machine

AutoHijacker: Automatic Indirect Prompt Injection Against Black-box LLM Agents