Empowering LLMs in Task-Oriented Dialogues: A Domain-Independent Multi-Agent Framework and Fine-Tuning Strategy

TL;DR

This paper introduces a domain-independent multi-agent framework for task-oriented dialogue systems based on LLMs, improving generalization and performance on lightweight models through specialized agents and novel training strategies.

Contribution

It proposes a multi-agent framework that separates dialogue tasks into domain-independent components and introduces DPO and DDA methods to enhance training and generalization.

Findings

Achieves better average performance on MultiWOZ datasets

Demonstrates strong generalizability and zero-shot capabilities

Enhances contextual understanding with DPO and DDA methods

Abstract

Task-oriented dialogue systems based on Large Language Models (LLMs) have gained increasing attention across various industries and achieved significant results. Current approaches condense complex procedural workflows into a single agent to achieve satisfactory performance on large-scale LLMs. However, these approaches face challenges to achieve comparable performance on fine-tuned lightweight LLMs, due to their limited capabilities in handling multiple complex logic. In this work, we design a Domain-Independent Multi-Agent Framework (DIMF), which contains Intent Classification Agent, Slot Filling Agent and Response Agent. This approach simplifies the learning complexity and enhances the generalization ability by separating the tasks into domain-independent components. In this framework, we enhance the capabilities in contextual understanding using the Direct Preference Optimisation (DPO) method, and propose a simple and effective Data Distribution Adaptation (DDA) method to mitigate degradation issues during DPO training. Experiments conducted on the MultiWOZ datasets show that our proposed method achieves a better average performance among all the baselines. Extensive analysis also demonstrates that our proposed framework exhibits excellent generalizability and zero-shot capability.