Multi Agent System
Multi-agent systems (MAS) research focuses on designing and analyzing systems composed of multiple interacting agents, aiming to achieve collective goals exceeding individual capabilities. Current research emphasizes efficient communication strategies within MAS, particularly leveraging large language models (LLMs) and incorporating techniques like Retrieval-Augmented Generation (RAG) to improve decision-making and reduce computational costs. This field is significant for advancing AI capabilities in complex problem-solving, with applications ranging from robotics and urban planning to financial modeling and software development. The development of robust and scalable frameworks, along with methods for handling malicious agents and model uncertainty, are key areas of ongoing investigation.
Papers - Page 5
ScoreFlow: Mastering LLM Agent Workflows via Score-based Preference Optimization
Free Energy Risk Metrics for Systemically Safe AI: Gatekeeping Multi-Agent Study
Multi-agent Architecture Search via Agentic Supernet
PAGNet: Pluggable Adaptive Generative Networks for Information Completion in Multi-Agent Communication
Large Language Models for Multi-Robot Systems: A Survey