DyTopo: Dynamic Topology Routing for Multi-Agent Reasoning via Semantic Matching

2026-02-05

AI Summary

Existing multi-agent systems typically employ fixed communication topologies during multi-round reasoning, which struggle to adapt to the dynamic collaboration requirements across different reasoning stages. This work proposes DyTopo, a novel framework that, for the first time, enables dynamic reconstruction of the communication topology at each reasoning round. Specifically, a manager constructs a sparse directed communication graph by matching semantic embeddings of natural language query and supply descriptors provided by agents, thereby enabling on-demand and interpretable message routing. Integrating large language models with lightweight semantic matching, DyTopo consistently outperforms the strongest baseline across four mainstream LLM backbones on code generation and mathematical reasoning tasks, achieving an average performance gain of 6.2% while yielding interpretable collaboration trajectories.

Technology Category

• Multiagent Systems: Agent Communication
• Planning, Routing, and Scheduling: Planning with Language Models
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Multi-agent systems built from prompted large language models can improve multi-round reasoning, yet most existing pipelines rely on fixed, trajectory-wide communication patterns that are poorly matched to the stage-dependent needs of iterative problem solving. We introduce DyTopo, a manager-guided multi-agent framework that reconstructs a sparse directed communication graph at each round. Conditioned on the manager's round goal, each agent outputs lightweight natural-language query (need) and \key (offer) descriptors; DyTopo embeds these descriptors and performs semantic matching, routing private messages only along the induced edges. Across code generation and mathematical reasoning benchmarks and four LLM backbones, DyTopo consistently outperforms over the strongest baseline (avg. +6.2). Beyond accuracy, DyTopo yields an interpretable coordination trace via the evolving graphs, enabling qualitative inspection of how communication pathways reconfigure across rounds.

Problem

• multi-agent systems
• dynamic topology
• communication routing
• iterative reasoning
• semantic matching

Innovation

• Dynamic Topology
• Semantic Matching
• Multi-Agent Reasoning
• Sparse Communication Graph
• LLM-based Agents