1 citations · 1 across the 9 of their papers we have counts for
11 papers
Representation Transitions Reveal Emerging Safety Risks in Multi-Turn LLM Agents
Haoyu Wang, Wei Zhao, Yedi Zhang +2
Multi-turn attacks on agentic systems can compose individually permissible actions into harmful outcomes, challenging defenses that assess actions or states in isolation. We show t…
Not All Tokens Are Equal: Region-Aware Consistency Repair of Backdoors in MLLMs
Jiali Wei, Ming Fan, Mingkun Zhang +6
MLLMs are increasingly deployed in user-facing applications, yet they inherit backdoor risks from the pipelines used to construct them: triggers may reside in images, texts, or bot…
MemGuard: Persisting Verifier Signals for LLM-Agent Memory Governance
Haoyu Wang, Guangyuan Dong, He Liang +5
LLM agents are moving from single-prompt use to long task streams in which reusable memory becomes a core capability for terminal, software-engineering, and web tasks. Such memory…
: Improving Agent Safety through Multi-Stage Defense
Zibo Xiao, Haoyu Wang, Jun Sun
Large Language Model (LLM) agents rely on multi-stage agentic workflows, with stages such as memory, planning, and tool execution, to accomplish complex tasks. However, risks may e…
SafeClaw-R: Towards Safe and Secure Multi-Agent Personal Assistants
Haoyu Wang, Zibo Xiao, Yedi Zhang +2
LLM-based multi-agent systems (MASs) are transforming personal productivity by autonomously executing complex, cross-platform tasks. Frameworks such as OpenClaw demonstrate the pot…
Domain-Specialized Tree of Thought through Plug-and-Play Predictors
Xuanqi Gao, Haoyu Wang, Jun Sun +2
While Large Language Models (LLMs) have advanced complex reasoning, prominent methods like the Tree of Thoughts (ToT) framework face a critical trade-off between exploration depth…
Robust and Efficient Tool Orchestration via Layered Execution Structures with Reflective Correction
Tao Zhe, Haoyu Wang, Bo Luo +6
Tool invocation is a core capability of agentic systems, yet failures often arise not from individual tool calls but from how multiple tools are organized and executed together. Ex…
LLM-enabled Applications Require System-Level Threat Monitoring
Yedi Zhang, Haoyu Wang, Xianglin Yang +2
LLM-enabled applications are rapidly reshaping the software ecosystem by using large language models as core reasoning components for complex task execution. This paradigm shift, h…
ProbGuard: Proactive Runtime Monitoring for LLM Agent Safety via Probabilistic Prediction
Haoyu Wang, Christopher M. Poskitt, Jiali Wei +1
Large Language Model (LLM) agents increasingly operate across domains such as robotics, virtual assistants, and web automation. However, their stochastic decision-making introduces…
AgentSpec: Customizable Runtime Enforcement for Safe and Reliable LLM Agents
Haoyu Wang, Christopher M. Poskitt, Jun Sun
Agents built on LLMs are increasingly deployed across diverse domains, automating complex decision-making and task execution. However, their autonomy introduces safety risks, inclu…
Evaluating and Mitigating Linguistic Discrimination in Large Language Models
Guoliang Dong, Haoyu Wang, Jun Sun +1
By training on text in various languages, large language models (LLMs) typically possess multilingual support and demonstrate remarkable capabilities in solving tasks described in…