AgentRouter: A Knowledge-Graph-Guided LLM Router for Collaborative Multi-Agent Question Answering
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866915535738896384 |
|---|---|
| author | Zhang, Zheyuan Shi, Kaiwen Yuan, Zhengqing Wang, Zehong Ma, Tianyi Murugesan, Keerthiram Galassi, Vincent Zhang, Chuxu Ye, Yanfang |
| author_facet | Zhang, Zheyuan Shi, Kaiwen Yuan, Zhengqing Wang, Zehong Ma, Tianyi Murugesan, Keerthiram Galassi, Vincent Zhang, Chuxu Ye, Yanfang |
| contents | Large language models (LLMs) and agent-based frameworks have advanced rapidly, enabling diverse applications. Yet, with the proliferation of models and agentic strategies, practitioners face substantial uncertainty in selecting the best configuration for a downstream task. Prior studies show that different agents and backbones exhibit complementary strengths, and that larger models are not always superior, underscoring the need for adaptive routing mechanisms. Existing approaches to agent routing, however, often emphasize cost efficiency while overlooking the fine-grained contextual and relational structure inherent in QA tasks. In this paper, we propose tAgentRouter, a framework that formulates multi-agent QA as a knowledge-graph-guided routing problem supervised by empirical performance signals. Specifically, we convert QA instance into a knowledge graph that jointly encodes queries, contextual entities, and agents, and then train a heterogeneous graph neural network (GNN) to propagate information across node types and produce task-aware routing distributions over agents. By leveraging soft supervision and weighted aggregation of agent outputs, AgentRouter learns principled collaboration schemes that capture the complementary strengths of diverse agents. Extensive experiments demonstrate that our framework consistently outperforms single-agent and ensemble baselines, while generalizing across benchmarks and LLM backbones. These results highlight the effectiveness and robustness of graph-supervised multi-agent routing for question answering. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_05445 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | AgentRouter: A Knowledge-Graph-Guided LLM Router for Collaborative Multi-Agent Question Answering Zhang, Zheyuan Shi, Kaiwen Yuan, Zhengqing Wang, Zehong Ma, Tianyi Murugesan, Keerthiram Galassi, Vincent Zhang, Chuxu Ye, Yanfang Computation and Language Large language models (LLMs) and agent-based frameworks have advanced rapidly, enabling diverse applications. Yet, with the proliferation of models and agentic strategies, practitioners face substantial uncertainty in selecting the best configuration for a downstream task. Prior studies show that different agents and backbones exhibit complementary strengths, and that larger models are not always superior, underscoring the need for adaptive routing mechanisms. Existing approaches to agent routing, however, often emphasize cost efficiency while overlooking the fine-grained contextual and relational structure inherent in QA tasks. In this paper, we propose tAgentRouter, a framework that formulates multi-agent QA as a knowledge-graph-guided routing problem supervised by empirical performance signals. Specifically, we convert QA instance into a knowledge graph that jointly encodes queries, contextual entities, and agents, and then train a heterogeneous graph neural network (GNN) to propagate information across node types and produce task-aware routing distributions over agents. By leveraging soft supervision and weighted aggregation of agent outputs, AgentRouter learns principled collaboration schemes that capture the complementary strengths of diverse agents. Extensive experiments demonstrate that our framework consistently outperforms single-agent and ensemble baselines, while generalizing across benchmarks and LLM backbones. These results highlight the effectiveness and robustness of graph-supervised multi-agent routing for question answering. |
| title | AgentRouter: A Knowledge-Graph-Guided LLM Router for Collaborative Multi-Agent Question Answering |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2510.05445 |