Multi-agent Application System in Office Collaboration Scenarios

TL;DR

This work presents a multi-agent application system for office collaboration that integrates AI, ML, and NLP to improve task allocation, progress monitoring, and information sharing. It introduces a Planner+Solver architecture within a master-slave multi-agent framework, augmented by multi-turn query rewriting and tool recall to support multi-intent dialogue and robust tool usage. A comprehensive data-engineering pipeline, including a multi-turn data generation framework and a Label Studio-based annotation scheme, underpins training and evaluation across query rewriting, tool recall, planning, and solving. Validated in real business settings, the system demonstrates strong performance in query understanding, task planning, and tool invocation, with WPS365 Assistant and AI Ask deployed to enhance real-world office workflows. The approach targets scalable, complex interactions in dynamic environments and large-scale multi-agent systems, highlighting practical impact for enterprise collaboration.

Abstract

This paper introduces a multi-agent application system designed to enhance office collaboration efficiency and work quality. The system integrates artificial intelligence, machine learning, and natural language processing technologies, achieving functionalities such as task allocation, progress monitoring, and information sharing. The agents within the system are capable of providing personalized collaboration support based on team members' needs and incorporate data analysis tools to improve decision-making quality. The paper also proposes an intelligent agent architecture that separates Plan and Solver, and through techniques such as multi-turn query rewriting and business tool retrieval, it enhances the agent's multi-intent and multi-turn dialogue capabilities. Furthermore, the paper details the design of tools and multi-turn dialogue in the context of office collaboration scenarios, and validates the system's effectiveness through experiments and evaluations. Ultimately, the system has demonstrated outstanding performance in real business applications, particularly in query understanding, task planning, and tool calling. Looking forward, the system is expected to play a more significant role in addressing complex interaction issues within dynamic environments and large-scale multi-agent systems.

Figures (6)

• Figure 1: A Visualization of Single and Multi-agent Architectures masterman2024landscape.
• Figure 2: System Architecture.
• Figure 3: Planner and Solver Flowchart.
• Figure 4: Query Rewrite Flowchart.
• Figure 5: Data Annotation Interface ().