4 citations · 12 across the 20 of their papers we have counts for
21 papers
Reassembling Distributed Risk: Trajectory-Conditioned Action Generation for Multi-Turn Agent Safety
Yanbo Dai, Zhenlan Ji, Zongjie Li +1
Tool-using LLM agents extend security risks beyond generated text to actions that affect external systems. Under multi-turn decomposition attacks, a harmful objective can be distri…
Detecting and Understanding Vulnerabilities in Fully Homomorphic Encryption Frameworks
Yiteng Peng, Dongwei Xiao, Zhibo Liu +2
Fully homomorphic encryption (FHE) allows computations to be performed directly on encrypted data without decryption, offering strong privacy guarantees for sensitive data analysis…
Beyond Content Safety: Real-Time Monitoring for Reasoning Vulnerabilities in Large Language Models
Xunguang Wang, Yuguang Zhou, Qingyue Wang +5
Large language models increasingly rely on explicit chain-of-thought reasoning to solve complex tasks, yet the safety of the reasoning process itself remains largely unaddressed. E…
CAM: A Causality-based Analysis Framework for Multi-Agent Code Generation Systems
Zongyi Lyu, Zhenlan Ji, Songqiang Chen +4
Despite the remarkable success that Multi-Agent Code Generation Systems (MACGS) have achieved, the inherent complexity of multi-agent architectures produces substantial volumes of…
SEAL: Subspace-Anchored Watermarks for LLM Ownership
Yanbo Dai, Zongjie Li, Zhenlan Ji +1
Large language models (LLMs) have achieved remarkable success across a wide range of natural language processing tasks, demonstrating human-level performance in text generation, re…
Understanding and Bridging the Planner-Coder Gap: A Systematic Study on the Robustness of Multi-Agent Systems for Code Generation
Zongyi Lyu, Songqiang Chen, Zhenlan Ji +5
Multi-agent systems (MASs) have emerged as a promising paradigm for automated code generation, demonstrating impressive performance on established benchmarks. Despite their prosper…
A Good Plan is Hard to Find: Aligning Models with Preferences is Misaligned with What Helps Users
Nishant Balepur, Matthew Shu, Yoo Yeon Sung +5
To assist users in complex tasks, LLMs generate plans: step-by-step instructions towards a goal. While alignment methods aim to ensure LLM plans are helpful, they train (RLHF) or e…
Digging Into the Internal: Causality-Based Analysis of LLM Function Calling
Zhenlan Ji, Daoyuan Wu, Wenxuan Wang +3
Function calling (FC) has emerged as a powerful technique for facilitating large language models (LLMs) to interact with external systems and perform structured tasks. However, the…
DisarmRAG: Stealthy Retriever-Centric Poisoning to Disable Self-Correction in Retrieval-Augmented Generation (Extended Version)
Yanbo Dai, Zhenlan Ji, Zongjie Li +2
Retrieval-Augmented Generation (RAG) has become a standard approach for improving the reliability of large language models (LLMs). Prior work demonstrates the vulnerability of RAG…
Evaluating LLMs on Sequential API Call Through Automated Test Generation
Yuheng Huang, Jiayang Song, Da Song +4
By integrating tools from external APIs, Large Language Models (LLMs) have expanded their promising capabilities in a diverse spectrum of complex real-world tasks. However, testing…
IP Leakage Attacks Targeting LLM-Based Multi-Agent Systems
Liwen Wang, Wenxuan Wang, Shuai Wang +5
The rapid advancement of Large Language Models (LLMs) has led to the emergence of Multi-Agent Systems (MAS) to perform complex tasks through collaboration. However, the intricate n…
SoK: Evaluating Jailbreak Guardrails for Large Language Models
Xunguang Wang, Zhenlan Ji, Wenxuan Wang +3
Large Language Models (LLMs) have achieved remarkable progress, but their deployment has exposed critical vulnerabilities, particularly to jailbreak attacks that circumvent safety…
EAMET: Robust Massive Model Editing via Embedding Alignment Optimization
Yanbo Dai, Zhenlan Ji, Zongjie Li +1
Model editing techniques are essential for efficiently updating knowledge in large language models (LLMs). However, the effectiveness of existing approaches degrades in massive edi…
STShield: Single-Token Sentinel for Real-Time Jailbreak Detection in Large Language Models
Xunguang Wang, Wenxuan Wang, Zhenlan Ji +4
Large Language Models (LLMs) have become increasingly vulnerable to jailbreak attacks that circumvent their safety mechanisms. While existing defense methods either suffer from ada…
SelfDefend: LLMs Can Defend Themselves against Jailbreaking in a Practical Manner
Xunguang Wang, Daoyuan Wu, Zhenlan Ji +7
Jailbreaking is an emerging adversarial attack that bypasses the safety alignment deployed in off-the-shelf large language models (LLMs) and has evolved into multiple categories: h…
Testing and Understanding Erroneous Planning in LLM Agents through Synthesized User Inputs
Zhenlan Ji, Daoyuan Wu, Pingchuan Ma +2
Agents based on large language models (LLMs) have demonstrated effectiveness in solving a wide range of tasks by integrating LLMs with key modules such as planning, memory, and too…
InstructTA: Instruction-Tuned Targeted Attack for Large Vision-Language Models
Xunguang Wang, Zhenlan Ji, Pingchuan Ma +3
Large vision-language models (LVLMs) have demonstrated their incredible capability in visual question answering. However, this rich visual interaction also makes LVLMs vulnerable t…
Benchmarking and Explaining Large Language Model-based Code Generation: A Causality-Centric Approach
Zhenlan Ji, Pingchuan Ma, Zongjie Li +1
While code generation has been widely used in various software development scenarios, the quality of the generated code is not guaranteed. This has been a particular concern in the…
Enabling Runtime Verification of Causal Discovery Algorithms with Automated Conditional Independence Reasoning (Extended Version)
Pingchuan Ma, Zhenlan Ji, Peisen Yao +2
Causal discovery is a powerful technique for identifying causal relationships among variables in data. It has been widely used in various applications in software engineering. Caus…
Causality-Aided Trade-off Analysis for Machine Learning Fairness
Zhenlan Ji, Pingchuan Ma, Shuai Wang +1
There has been an increasing interest in enhancing the fairness of machine learning (ML). Despite the growing number of fairness-improving methods, we lack a systematic understandi…
PerfCE: Performance Debugging on Databases with Chaos Engineering-Enhanced Causality Analysis
Zhenlan Ji, Pingchuan Ma, Shuai Wang
Debugging performance anomalies in real-world databases is challenging. Causal inference techniques enable qualitative and quantitative root cause analysis of performance downgrade…