1 citations · 3 across the 16 of their papers we have counts for
20 papers
When Reasoning Goes Astray: Attention Dynamics of Uncontrolled Reasoning
Yuanhe Zhang, Ziwei Wang, Jie Ren +6
Large reasoning models (LRMs) improve performance on complex tasks through extended reasoning, yet the same process can degenerate into redundant verification and persistent genera…
Uncovering Uncontrolled Repetition through Residual Stream Dynamics
Yuanhe Zhang, Xinyao Zhou, Haoran Gao +5
Uncontrolled repetition can prolong autoregressive generation in large language models (LLMs) and enable resource consumption attacks. Prior analyses of repetitive generation have…
Long-Term Memory-Guided Enhancement for Target Perception in Audio-Language Models
Zhenhong Zhou, Xuanyue Zhao, Youji Liu +4
Audio large language models (ALLMs) can reason about the content of audio recordings to perform complex tasks. However, these capabilities usually collapse in real-world environmen…
Resource Consumption Threats in Large Language Models
Yuanhe Zhang, Xinyue Wang, Zhican Chen +8
Given limited and costly computational infrastructure, resource efficiency is a key requirement for large language models (LLMs). Efficient LLMs increase service capacity for provi…
From Helpfulness to Toxic Proactivity: Diagnosing Behavioral Misalignment in LLM Agents
Xinyue Wang, Yuanhe Zhang, Zhengshuo Gong +6
The enhanced capabilities of LLM-based agents come with an emergency for model planning and tool-use abilities. Attributing to helpful-harmless trade-off from LLM alignment, agents…
Memory in the Age of AI Agents
Yuyang Hu, Shichun Liu, Yanwei Yue +44
Memory has emerged, and will continue to remain, a core capability of foundation model-based agents. As research on agent memory rapidly expands and attracts unprecedented attentio…
SEE: Signal Embedding Energy for Quantifying Noise Interference in Large Audio Language Models
Yuanhe Zhang, Jiayu Tian, Yibo Zhang +5
Large Audio Language Models (LALMs) have been widely applied in real-time scenarios, such as in-car assistants and online meeting comprehension. In practice, audio inputs are often…
LeechHijack: Covert Computational Resource Exploitation in Intelligent Agent Systems
Yuanhe Zhang, Weiliu Wang, Zhenhong Zhou +5
Large Language Model (LLM)-based agents have demonstrated remarkable capabilities in reasoning, planning, and tool usage. The recently proposed Model Context Protocol (MCP) has eme…
LIFEBench: Evaluating Length Instruction Following in Large Language Models
Wei Zhang, Zhenhong Zhou, Kun Wang +9
While large language models (LLMs) can solve PhD-level reasoning problems over long context inputs, they still struggle with a seemingly simpler task: following explicit length ins…
: A Pluggable and Dynamic DoS-Defense Framework Against Resource Consumption Attacks Targeting Large Language Models
Yuanhe Zhang, Xinyue Wang, Haoran Gao +4
Large Language Models (LLMs), due to substantial computational requirements, are vulnerable to resource consumption attacks, which can severely degrade server performance or even c…
A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deployment
Kun Wang, Guibin Zhang, Zhenhong Zhou +100
The remarkable success of Large Language Models (LLMs) has illuminated a promising pathway toward achieving Artificial General Intelligence for both academic and industrial communi…
CORBA: Contagious Recursive Blocking Attacks on Multi-Agent Systems Based on Large Language Models
Zhenhong Zhou, Zherui Li, Jie Zhang +4
Large Language Model-based Multi-Agent Systems (LLM-MASs) have demonstrated remarkable real-world capabilities, effectively collaborating to complete complex tasks. While these sys…
DemonAgent: Dynamically Encrypted Multi-Backdoor Implantation Attack on LLM-based Agent
Pengyu Zhu, Zhenhong Zhou, Yuanhe Zhang +3
As LLM-based agents become increasingly prevalent, backdoors can be implanted into agents through user queries or environment feedback, raising critical concerns regarding safety v…
Crabs: Consuming Resource via Auto-generation for LLM-DoS Attack under Black-box Settings
Yuanhe Zhang, Zhenhong Zhou, Wei Zhang +4
Large Language Models (LLMs) have demonstrated remarkable performance across diverse tasks yet still are vulnerable to external threats, particularly LLM Denial-of-Service (LLM-DoS…
On the Role of Attention Heads in Large Language Model Safety
Zhenhong Zhou, Haiyang Yu, Xinghua Zhang +6
Large language models (LLMs) achieve state-of-the-art performance on multiple language tasks, yet their safety guardrails can be circumvented, leading to harmful generations. In li…
Alignment-Enhanced Decoding:Defending via Token-Level Adaptive Refining of Probability Distributions
Quan Liu, Zhenhong Zhou, Longzhu He +3
Large language models are susceptible to jailbreak attacks, which can result in the generation of harmful content. While prior defenses mitigate these risks by perturbing or inspec…
Course-Correction: Safety Alignment Using Synthetic Preferences
Rongwu Xu, Yishuo Cai, Zhenhong Zhou +6
The risk of harmful content generated by large language models (LLMs) becomes a critical concern. This paper presents a systematic study on assessing and improving LLMs' capability…
How Alignment and Jailbreak Work: Explain LLM Safety through Intermediate Hidden States
Zhenhong Zhou, Haiyang Yu, Xinghua Zhang +3
Large language models (LLMs) rely on safety alignment to avoid responding to malicious user inputs. Unfortunately, jailbreak can circumvent safety guardrails, resulting in LLMs gen…
Speak Out of Turn: Safety Vulnerability of Large Language Models in Multi-turn Dialogue
Zhenhong Zhou, Jiuyang Xiang, Haopeng Chen +3
Large Language Models (LLMs) have been demonstrated to generate illegal or unethical responses, particularly when subjected to "jailbreak." Research on jailbreak has highlighted th…
Quantifying and Analyzing Entity-level Memorization in Large Language Models
Zhenhong Zhou, Jiuyang Xiang, Chaomeng Chen +1
Large language models (LLMs) have been proven capable of memorizing their training data, which can be extracted through specifically designed prompts. As the scale of datasets cont…