5 papers
Diagnosis-Driven Automatic Repair for Agentic Workflow via Symbolic Inference
Xuyan Ma, Yawen Wang, Junjie Wang +5
Platform-orchestrated agentic workflows have become a popular paradigm for developing LLM-based applications. However, their reliability remains a major challenge due to the uncert…
Where Did It Go Wrong? Capability-Oriented Failure Attribution for Vision-and-Language Navigation Agents
Jianming Chen, Yawen Wang, Junjie Wang +4
Embodied agents in safety-critical applications such as Vision-Language Navigation (VLN) rely on multiple interdependent capabilities (e.g., perception, memory, planning, decision)…
Demystifying the Lifecycle of Failures in Platform-Orchestrated Agentic Workflows
Xuyan Ma, Xiaofei Xie, Yawen Wang +4
Agentic workflows built on low-code orchestration platforms enable rapid development of multi-agent systems, but they also introduce new and poorly understood failure modes that hi…
Adversarial Attack on Black-Box Multi-Agent by Adaptive Perturbation
Jianming Chen, Yawen Wang, Junjie Wang +4
Evaluating security and reliability for multi-agent systems (MAS) is urgent as they become increasingly prevalent in various applications. As an evaluation technique, existing adve…
Robust Multi-agent Communication Based on Decentralization-Oriented Adversarial Training
Xuyan Ma, Yawen Wang, Junjie Wang +5
In typical multi-agent reinforcement learning (MARL) problems, communication is important for agents to share information and make the right decisions. However, due to the complexi…