Recent advancements in multi-agent systems are focusing on optimizing performance and efficiency in real-world applications, particularly in environments constrained by latency and communication bandwidth. New frameworks are being developed to enhance orchestration and coordination among agents, such as latency-aware orchestration that reduces critical execution paths by nearly 46%, and bandwidth-efficient communication strategies that improve performance by over 180% while cutting bandwidth usage by 41%. These innovations are crucial for deploying multi-agent systems in robotics, autonomous vehicles, and distributed networks, where timely and effective communication is essential. Additionally, the integration of structured debate mechanisms and differentiable modal logic is paving the way for more transparent decision-making processes, allowing for better handling of complex scenarios in fields like criminal justice and healthcare. Overall, the field is moving toward more adaptive, robust, and efficient systems capable of addressing the intricacies of real-world tasks while maintaining high levels of performance and reliability.
Despite rapid developments and widespread applications of MLLM agents, they still struggle with long-form video understanding (LVU) tasks, which are characterized by high information density and exten...
Multi-Agent Debate (MAD) has emerged as a promising paradigm for enhancing large language model reasoning. However, recent work reveals a limitation:standard MAD cannot improve belief correctness beyo...
We introduce AgenticSimLaw, a role-structured, multi-agent debate framework that provides transparent and controllable test-time reasoning for high-stakes tabular decision-making tasks. Unlike black-b...
Multi-agent LLM orchestration incurs synchronization costs scaling as O(n x S x |D|) in agents, steps, and artifact size under naive broadcast -- a regime I term broadcast-induced triply-multiplicativ...
Multi-agent reinforcement learning systems deployed in real-world robotics applications face severe communication constraints that significantly impact coordination effectiveness. We present a framewo...
Most existing Large Language Model (LLM)-based Multi-Agent Systems (MAS) rely on predefined workflows, where human engineers enumerate task states in advance and specify routing rules and contextual i...
Multi-agent systems (MAS) enable complex reasoning by coordinating multiple agents, but often incur high inference latency due to multi-step execution and repeated model invocations, severely limiting...
Multi-agent coordination dilemmas expose a fundamental tension between individual optimization and collective welfare, yet characterizing such coordination requires metrics sensitive to temporal struc...
Large Language Model-based Multi-Agent Systems (LLM-based MAS), where multiple LLM agents collaborate to solve complex tasks, have shown impressive performance in many areas. However, MAS are typicall...
Large Language Models (LLMs) have demonstrated remarkable capabilities across a wide range of language tasks, yet complex multi-step reasoning remains a fundamental challenge. While Large Reasoning Mo...