Triad: A Framework Leveraging a Multi-Role LLM-based Agent to Solve Knowledge Base Question Answering

TL;DR

Triad introduces a multi-role LLM-based agent framework for knowledge base question answering, effectively addressing data scarcity and task complexity, and outperforming existing systems on benchmark datasets.

Contribution

The paper proposes a novel multi-role agent framework, Triad, for KBQA that leverages LLMs to improve performance without extensive task-specific training.

Findings

Outperforms state-of-the-art on LC-QuAD and YAGO-QA benchmarks.

Achieves F1 scores of 11.8% and 20.7% higher than previous methods.

Demonstrates effective collaboration of multiple agent roles in KBQA tasks.

Abstract

Recent progress with LLM-based agents has shown promising results across various tasks. However, their use in answering questions from knowledge bases remains largely unexplored. Implementing a KBQA system using traditional methods is challenging due to the shortage of task-specific training data and the complexity of creating task-focused model structures. In this paper, we present Triad, a unified framework that utilizes an LLM-based agent with three roles for KBQA tasks. The agent is assigned three roles to tackle different KBQA subtasks: agent as a generalist for mastering various subtasks, as a decision maker for the selection of candidates, and as an advisor for answering questions with knowledge. Our KBQA framework is executed in four phases, involving the collaboration of the agent's multiple roles. We evaluated the performance of our framework using three benchmark datasets, and the results show that our framework outperforms state-of-the-art systems on the LC-QuAD and YAGO-QA benchmarks, yielding F1 scores of 11.8% and 20.7%, respectively.