449 citations · 1.7k across the 108 of their papers we have counts for
113 papers
agent-risk: Risk Assessment for Changes to AI Agents
Yue Zhao, Zixiang Xu, Angela Huyue Zhang +2
Before changing a deployed AI agent (a stricter safety gate, a new model), organizations must approve the change, starting from existing logs. Extending classical risk analysis to…
MetaPersona: Task-Grounded Synthetic Populations from Empirical Social Science
Jinyi Ye, Yuangang Li, Chenxiao Yu +7
Personas used to seed LLM social simulations face a cold-start problem: existing methods lack a principled basis for deciding which attributes to include and how to assign their va…
Controlled Decoding Attacks on Black-Box LLMs
Jesson Wang, Shawn Li, Wei Yang +4
Manipulating next-token probabilities during generation can bypass the safety alignment of large language models. Existing approaches, however, rely on access to model weights or n…
CatchBench: When Can an Agent Failure Be Caught?
Yue Zhao, Mengyuan Li, Ruolin Li +5
When can an agent failure be caught? A weak audit score alone cannot identify whether the record or the method is limiting. CatchBench therefore puts one auditor's question to thre…
Beyond Top- Skill Retrieval: Diversity-Aware Skill Routing for LLM Agents
Wang Wei, Tiankai Yang, Samyadeep Basu +7
Large language model (LLM) agents increasingly rely on external skills, but routing user requests over large skill registries is difficult because many skills are functionally redu…
WeClawArena: An Auditable Sandbox and Benchmark for Cross-User Agents Collaboration and Security in Human-Centered Agent Networks
Prince Zizhuang Wang, Aojie Yuan, Haiyue Zhang +3
Recent advances in persistent personal-agent frameworks are making human-centered agent networks realistic deployment targets: each user can be served by an AI agent that acts on t…
JigShape: Evaluating Visual-Geometric Reasoning in VLMs through Jigsaw Puzzles
Shawn Li, Wei Yang, Jike Zhong +11
Jigsaw puzzle solving requires jointly reasoning about visual content and geometric constraints, yet existing benchmarks use rectangular cuts that create ambiguous ground truth in…
MemoHarness: Agent Harnesses That Learn from Experience
Yue Huang, Wenjie Wang, Han Bao +7
An agent harness is the external control layer that turns a base LLM into an executable agent by managing context, tools, orchestration, memory, decoding, and output handling. Whil…
SkillCenter: A Large-Scale Source-Grounded Skill Library for Autonomous AI Agents
Tianming Sha, Yue Zhao, Lichao Sun +1
Autonomous AI agents can execute complex tasks with limited human review, yet they often lack the grounded operational knowledge to make their outputs not just executable but corre…
SEVA: Self-Evolving Verification Agent with Process Reward for Fact Attribution
Aojie Yuan, Yi Nian, Haiyue Zhang +2
Hallucination is the reliability bottleneck for LLM-based agents, and fact attribution verifiers are the last line of defense -- yet today's verifiers emit only opaque binary label…
Agent Safety Is Action Alignment
Shawn Li, Yue Zhao
Large language models increasingly act as agents: they call tools, move money, delete records, and send messages on a user's behalf. To keep them safe, practitioners imported the c…
GRADE: Graph Representation of LLM Agent Dependency and Execution
Yue Zhao
Traces record what LLM agents did, leaving implicit what each step relied on. GRADE holds both in one graph; six published graph views are its projections. Execution edges come fre…
FORTIS: Benchmarking Over-Privilege in Agent Skills
Shawn Li, Chenxiao Yu, Han Wang +8
Large language model agents increasingly operate through an intermediate skill layer that mediates between user intent and concrete task execution. This layer is widely treated as…
Memory Retrieval for Changing Preferences
Yuehan Qin, Li Li, Linxin Song +4
Long-context dialogue systems must decide both when to access memory and which parts of the interaction history are relevant. Existing approaches typically rely on heuristic retrie…
SpecAlign: Efficient Specification-Grounded Alignment of Large Language Models via Synthetic Data
Wenjie Wang, Yue Huang, Zhengqing Yuan +6
As large language models (LLMs) are increasingly deployed in real-world applications, alignment is no longer governed by a single universal notion of safety or helpfulness, but ins…
DOG-DPO:Dynamic Optimization in Geometry for Safety Alignment
Yi Nian, Tiankai Yang, Yudi Zhang +7
Safety alignment for large language models relies on preference data, but current pipelines often train on large, redundant datasets. Existing data selection methods typically scor…
MaskForge: Structure-Aware Adaptive Attacks for Jailbreaking Diffusion Large Language Models
Yingzi Ma, Zhengyue Zhao, Xiaogeng Liu +3
Diffusion large language models (dLLMs) generate text by iteratively denoising partially masked sequences under bidirectional context, exposing a safety surface distinct from autor…
GEO-Bench: Benchmarking Ranking Manipulation in Generative Engine Optimization
Ojas Nimase, Zhe Chen, Gengpei Qi +2
Large language models (LLMs) increasingly rank products, documents, and recommendations for user queries, which makes manipulating these rankings a growing concern for fairness and…
CHI-Bench: Can AI Agents Automate End-to-End, Long-Horizon, Policy-Rich Healthcare Workflows?
Haolin Chen, Deon Metelski, Leon Qi +30
End-to-end automation of realistic healthcare operations stresses three capabilities underrepresented in current benchmarks: policy density, decisions must be grounded in a large l…
Geometry over Density: Few-Shot Cross-Domain OOD Detection
Shawn Li, You Qin, Jiate Li +4
Out-of-distribution (OOD) detection identifies test samples that fall outside a model's training distribution, a capability critical for safe deployment in high-stakes applications…
MedExAgent: Training LLM Agents to Ask, Examine, and Diagnose in Noisy Clinical Environments
Yicheng Gao, Xiaolin Zhou, Yahan Li +2
Real-world clinical diagnosis is a complex process in which the doctor is required to obtain information from both interaction with the patient and conducting medical exams. Additi…
Can Subgraph Explanations Be Weaponized to Steal Graph Neural Networks?
Ojas Nimase, Jiate Li, Yue Zhao +1
Graph Machine Learning as a Service (GMLaaS) platforms increasingly implement explainability interfaces to meet regulatory transparency requirements. However, this transparency cre…
A Survey on LLM-based Conversational User Simulation
Bo Ni, Leyao Wang, Yu Wang +27
User simulation has long played a vital role in computer science due to its potential to support a wide range of applications. Language, as the primary medium of human communicatio…
CoAct: Co-Active LLM Preference Learning with Human-AI Synergy
Ruiyao Xu, Mihir Parmar, Tiankai Yang +3
Learning from preference-based feedback has become an effective approach for aligning LLMs across diverse tasks. However, high-quality human-annotated preference data remains expen…