Talk to Right Specialists: Iterative Routing in Multi-agent Systems for Question Answering

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wu, Feijie, Li, Zitao, Wei, Fei, Li, Yaliang, Ding, Bolin, Gao, Jing
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918428711845888
author Wu, Feijie
Li, Zitao
Wei, Fei
Li, Yaliang
Ding, Bolin
Gao, Jing
author_facet Wu, Feijie
Li, Zitao
Wei, Fei
Li, Yaliang
Ding, Bolin
Gao, Jing
contents Retrieval-augmented generation (RAG) agents are increasingly deployed to answer questions over local knowledge bases that cannot be centralized due to knowledge-sovereignty constraints. This results in two recurring failures in production: users do not know which agent to consult, and complex questions require evidence distributed across multiple agents. To overcome these challenges, we propose RIRS, a training-free orchestration framework to enable a multi-agent system for question answering. In detail, RIRS summarizes each agent's local corpus in an embedding space, enabling a user-facing server to route queries only to the most relevant agents, reducing latency and avoiding noisy "broadcast-to-all" contexts. For complicated questions, the server can iteratively aggregate responses to derive intermediate results and refine the question to bridge the gap toward a comprehensive answer. Extensive experiments demonstrate the effectiveness of RIRS, including its ability to precisely select agents and provide accurate responses to single-hop queries, and its use of an iterative strategy to achieve accurate, multi-step resolutions for complex queries.
format Preprint
id arxiv_https___arxiv_org_abs_2501_07813
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Talk to Right Specialists: Iterative Routing in Multi-agent Systems for Question Answering
Wu, Feijie
Li, Zitao
Wei, Fei
Li, Yaliang
Ding, Bolin
Gao, Jing
Multiagent Systems
Artificial Intelligence
Computation and Language
Retrieval-augmented generation (RAG) agents are increasingly deployed to answer questions over local knowledge bases that cannot be centralized due to knowledge-sovereignty constraints. This results in two recurring failures in production: users do not know which agent to consult, and complex questions require evidence distributed across multiple agents. To overcome these challenges, we propose RIRS, a training-free orchestration framework to enable a multi-agent system for question answering. In detail, RIRS summarizes each agent's local corpus in an embedding space, enabling a user-facing server to route queries only to the most relevant agents, reducing latency and avoiding noisy "broadcast-to-all" contexts. For complicated questions, the server can iteratively aggregate responses to derive intermediate results and refine the question to bridge the gap toward a comprehensive answer. Extensive experiments demonstrate the effectiveness of RIRS, including its ability to precisely select agents and provide accurate responses to single-hop queries, and its use of an iterative strategy to achieve accurate, multi-step resolutions for complex queries.
title Talk to Right Specialists: Iterative Routing in Multi-agent Systems for Question Answering
topic Multiagent Systems
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2501.07813