KIMAs: A Configurable Knowledge Integrated Multi-Agent System

TL;DR

KIMAs is a flexible, scalable multi-agent system designed to enhance knowledge-intensive applications with heterogeneous data, improved context management, and low-latency responses, facilitating real-world deployment of LLMs.

Contribution

This work introduces KIMAs, a configurable framework that integrates diverse knowledge sources with efficient retrieval, context handling, and parallel execution for practical LLM applications.

Findings

Successfully configured KIMAs for three real-world applications

Improved retrieval accuracy and conversational coherence

Achieved low-latency, scalable multi-agent deployment

Abstract

Knowledge-intensive conversations supported by large language models (LLMs) have become one of the most popular and helpful applications that can assist people in different aspects. Many current knowledge-intensive applications are centered on retrieval-augmented generation (RAG) techniques. While many open-source RAG frameworks facilitate the development of RAG-based applications, they often fall short in handling practical scenarios complicated by heterogeneous data in topics and formats, conversational context management, and the requirement of low-latency response times. This technical report presents a configurable knowledge integrated multi-agent system, KIMAs, to address these challenges. KIMAs features a flexible and configurable system for integrating diverse knowledge sources with 1) context management and query rewrite mechanisms to improve retrieval accuracy and multi-turn conversational coherency, 2) efficient knowledge routing and retrieval, 3) simple but effective filter and reference generation mechanisms, and 4) optimized parallelizable multi-agent pipeline execution. Our work provides a scalable framework for advancing the deployment of LLMs in real-world settings. To show how KIMAs can help developers build knowledge-intensive applications with different scales and emphases, we demonstrate how we configure the system to three applications already running in practice with reliable performance.