Yang Liu

AudioTrust: Benchmarking The Multifaceted Trustworthiness of Audio Large Language Models

MAS$^2$: Self-Generative, Self-Configuring, Self-Rectifying Multi-Agent Systems

MIDAS: Multi-Image Dispersion and Semantic Reconstruction for Jailbreaking MLLMs

SaFeR-VLM: Toward Safety-aware Fine-grained Reasoning in Multimodal Models

Text Modality Oriented Image Feature Extraction for Detecting Diffusion-based DeepFake

Obscure but Effective: Classical Chinese Jailbreak Prompt Optimization via Bio-Inspired Search

Resource Consumption Red-Teaming for Large Vision-Language Models

Inverse Reinforcement Learning with Dynamic Reward Scaling for LLM Alignment

Dual-Priv Pruning : Efficient Differential Private Fine-Tuning in Multimodal Large Language Models

Reasoning as an Attack Surface: Adaptive Evolutionary CoT Jailbreaks for LLMs

SAGE-CoT: Self-Adaptive Generated Chain-of-Thought for Jailbreaking

Learning Asymmetric Phase Dynamics via Clinically-Guided Spatiotemporal Fusion

CircuitProbe: Dissecting Spatiotemporal Visual Semantics with Circuit Tracing

On the Role of Attention Heads in Large Language Model Safety

DepthVanish: Optimizing Adversarial Interval Structures for Stereo-Depth-Invisible Patches

Adversarial Attacks against Closed-Source MLLMs via Feature Optimal Alignment

STAFF: Speculative Coreset Selection for Task-Specific Fine-tuning

Improved Techniques for Optimization-Based Jailbreaking on Large Language Models

TREANT: Red-teaming Text-to-Image Models with Tree-based Semantic Transformations

Impact-driven Context Filtering For Cross-file Code Completion

OT-Attack: Enhancing Adversarial Transferability of Vision-Language Models via Optimal Transport Optimization

SeCon-RAG: A Two-Stage Semantic Filtering and Conflict-Free Framework for Trustworthy RAG

REPOFILTER: Adaptive Retrieval Context Trimming for Repository-Level Code Completion

Defending LVLMs Against Vision Attacks Through Partial-Perception Supervision

MIND SCRAMBLE: UNVEILING LARGE LANGUAGE MODEL PSYCHOLOGY VIA TYPOGLYCEMIA