2 citations · 9 across the 19 of their papers we have counts for
22 papers
Omni-UBench: Benchmarking Omni-Modal LLM-based Guardrails Across Safety, Utility, and Reliability
Junhao Wang, Yidou Chen, Li Lu +6
Advances in Omni-modal Large Language Models (OLLMs) have enabled increasingly seamless human-AI interaction, but their unified processing of heterogeneous modalities also expands…
Lasso Games
Krzysztof Żyndul
We study games on finite graphs with labelled or weighted edges in which plays end as soon as the first loop is formed. The resulting play has a lasso shape: a path (head) followed…
Towards Merging-conditional Backdoor Unalignment against Large Language Models
Yukun Chen, Yidou Chen, Yiming Li +4
Model merging has become widely adopted for enhancing the task-specific performance of Large Language Models (LLMs) without additional training. While the specialized model from th…
ActBench: Self-Evolving Benchmark of Behavioral Safety in Cowork Agents
Hongwei Yao, Yiming Liu, Meihui Chen +6
Cowork agents may complete benign tasks while disclosing protected data, manipulating unauthorized state, invocate unauthorized API. We define behavioral safety and introduce ActBe…
Combating Knowledge Corruption in Agent Systems: A Byzantine-Tolerant Secure Collaborative RAG Framework
Zhaoqi Wang, Daqing He, Zijian Zhang +13
While retrieval-augmented generation systems partially address the hallucination issues in large language models, it also introduces new vulnerabilities to knowledge corruption att…
Red-Teaming Agent Execution Contexts: Open-World Security Evaluation on OpenClaw
Hongwei Yao, Yiming Liu, Yiling He +1
Agentic language-model systems increasingly rely on mutable execution contexts, including files, memory, tools, skills, and auxiliary artifacts, creating security risks beyond expl…
SRBench: A Comprehensive Benchmark for Sequential Recommendation with Large Language Models
Jianhong Li, Zeheng Qian, Wangze Ni +4
LLM development has aroused great interest in Sequential Recommendation (SR) applications. However, comprehensive evaluation of SR models remains lacking due to the limitations of…
AttriGuard: Defeating Indirect Prompt Injection in LLM Agents via Causal Attribution of Tool Invocations
Yu He, Haozhe Zhu, Yiming Li +4
LLM agents are highly vulnerable to Indirect Prompt Injection (IPI), where adversaries embed malicious directives in untrusted tool outputs to hijack execution. Most existing defen…
Black-Box Guardrail Reverse-engineering Attack
Hongwei Yao, Yun Xia, Shuo Shao +3
Large language models (LLMs) increasingly employ guardrails to enforce ethical, legal, and application-specific constraints on their outputs. While effective at mitigating harmful…
PromptCOS: Towards Content-only System Prompt Copyright Auditing for LLMs
Yuchen Yang, Yiming Li, Hongwei Yao +6
System prompts are critical for shaping the behavior and output quality of large language model (LLM)-based applications, driving substantial investment in optimizing high-quality…
Quantifying Conversation Drift in MCP via Latent Polytope
Haoran Shi, Hongwei Yao, Shuo Shao +4
The Model Context Protocol (MCP) enhances large language models (LLMs) by integrating external tools, enabling dynamic aggregation of real-time data to improve task execution. Howe…
SoK: Large Language Model Copyright Auditing via Fingerprinting
Shuo Shao, Yiming Li, Yu He +4
The broad capabilities and substantial resources required to train Large Language Models (LLMs) make them valuable intellectual property, yet they remain vulnerable to copyright in…
BadReward: Clean-Label Poisoning of Reward Models in Text-to-Image RLHF
Kaiwen Duan, Hongwei Yao, Yufei Chen +4
Reinforcement Learning from Human Feedback (RLHF) is crucial for aligning text-to-image (T2I) models with human preferences. However, RLHF's feedback mechanism also opens new pathw…
ControlNET: A Firewall for RAG-based LLM System
Hongwei Yao, Haoran Shi, Yidou Chen +3
Retrieval-Augmented Generation (RAG) has significantly enhanced the factual accuracy and domain adaptability of Large Language Models (LLMs). This advancement has enabled their wid…
FIT-Print: Towards False-claim-resistant Model Ownership Verification via Targeted Fingerprint
Shuo Shao, Haozhe Zhu, Yiming Li +3
Model fingerprinting has emerged as a crucial mechanism for safeguarding the intellectual property of open-source models, offering a non-intrusive approach that requires no modific…
Eguard: Defending LLM Embeddings Against Inversion Attacks via Text Mutual Information Optimization
Tiantian Liu, Hongwei Yao, Feng Lin +3
Embeddings have become a cornerstone in the functionality of large language models (LLMs) due to their ability to transform text data into rich, dense numerical representations tha…
ShadowCode: Towards (Automatic) External Prompt Injection Attack against Code LLMs
Yuchen Yang, Yiming Li, Hongwei Yao +5
Recent advancements have led to the widespread adoption of code-oriented large language models (Code LLMs) for programming tasks. Despite their success in deployment, their securit…
Explanation as a Watermark: Towards Harmless and Multi-bit Model Ownership Verification via Watermarking Feature Attribution
Shuo Shao, Yiming Li, Hongwei Yao +3
Ownership verification is currently the most critical and widely adopted post-hoc method to safeguard model copyright. In general, model owners exploit it to identify whether a giv…
PoisonPrompt: Backdoor Attack on Prompt-based Large Language Models
Hongwei Yao, Jian Lou, Zhan Qin
Prompts have significantly improved the performance of pretrained Large Language Models (LLMs) on various downstream tasks recently, making them increasingly indispensable for a di…
RemovalNet: DNN Fingerprint Removal Attacks
Hongwei Yao, Zheng Li, Kunzhe Huang +3
With the performance of deep neural networks (DNNs) remarkably improving, DNNs have been widely used in many areas. Consequently, the DNN model has become a valuable asset, and its…
PromptCARE: Prompt Copyright Protection by Watermark Injection and Verification
Hongwei Yao, Jian Lou, Kui Ren +1
Large language models (LLMs) have witnessed a meteoric rise in popularity among the general public users over the past few months, facilitating diverse downstream tasks with human-…
FDINet: Protecting against DNN Model Extraction via Feature Distortion Index
Hongwei Yao, Zheng Li, Haiqin Weng +3
Machine Learning as a Service (MLaaS) platforms have gained popularity due to their accessibility, cost-efficiency, scalability, and rapid development capabilities. However, recent…