TCAndon-Router: Adaptive Reasoning Router for Multi-Agent Collaboration

2026-01-08

arXiv.org

AI Summary

Existing task routing approaches rely on static single-label decisions, which struggle to accommodate dynamically joining agents and often lead to conflicts under capability overlap, thereby compromising the accuracy and robustness of multi-agent systems. This work proposes an adaptive reasoning-based routing mechanism that dynamically selects multiple candidate agents to collaboratively address a query by generating natural language reasoning chains, followed by a dedicated refinement agent that fuses their responses. Replacing static routing with multi-label reasoning, the method integrates dynamic agent registration, collaborative execution, and response refinement to significantly enhance routing accuracy and system robustness—particularly for complex or ambiguous queries—while also improving decision interpretability. Extensive experiments on both public benchmarks and real-world enterprise datasets validate the effectiveness of the proposed approach.

Technology Category

• Planning, Routing, and Scheduling: Model-Based Reasoning
• Multiagent Systems: Mechanism Design
• Intelligent Robots: Multi-Robot Systems

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Multi-Agent Systems(MAS) have become a powerful paradigm for building high performance intelligent applications. Within these systems, the router responsible for determining which expert agents should handle a given query plays a crucial role in overall performance. Existing routing strategies generally fall into two categories: performance routing, which balances latency and cost across models of different sizes, and task routing, which assigns queries to domain-specific experts to improve accuracy. In real-world enterprise applications, task routing is more suitable; however, most existing approaches rely on static single-label decisions, which introduce two major limitations: (i) difficulty in seamlessly integrating new agents as business domains expand, and (ii) routing conflicts caused by overlapping agent capabilities, ultimately degrading accuracy and robustness.To address these challenges, we propose TCAndon-Router(TCAR): an adaptive reasoning router for multi-agent collaboration. Unlike traditional routers, TCAR supports dynamic agent onboarding and first generates a natural-language reasoning chain before predicting a set of candidate agents capable of handling the query. In addition, we design a collaborative execution pipeline in which selected agents independently produce responses, which are then aggregated and refined into a single high-quality response by a dedicated Refining Agent.Experiments on public datasets and real enterprise data demonstrate that TCAR significantly improves routing accuracy, reduces routing conflicts, and remains robust in ambiguous scenarios. We have released TCAR at https://huggingface.co/tencent/TCAndon-Router to support future research on explainable and collaborative multi-agent routing.

Problem

• multi-agent systems
• task routing
• routing conflicts
• dynamic agent onboarding
• agent collaboration

Innovation

• adaptive reasoning router
• multi-agent collaboration
• dynamic agent onboarding
• reasoning chain
• collaborative execution pipeline