AgentRE: An Agent-Based Framework for Navigating Complex Information Landscapes in Relation Extraction

TL;DR

AgentRE introduces an agent-based framework utilizing large language models to improve relation extraction in complex, multilingual scenarios, especially effective in low-resource settings and capable of generating high-quality training data.

Contribution

This paper presents a novel agent-based relation extraction framework that leverages LLMs' memory, retrieval, and reflection to handle complex scenarios and generate training data.

Findings

Outperforms existing methods on English and Chinese datasets.

Effective in low-resource scenarios.

Can generate high-quality training datasets for smaller models.

Abstract

The relation extraction (RE) in complex scenarios faces challenges such as diverse relation types and ambiguous relations between entities within a single sentence, leading to the poor performance of pure "text-in, text-out" language models (LMs). To address these challenges, in this paper, we propose an agent-based RE framework, namely AgentRE, which fully leverages the potential of large language models (LLMs) including memory, retrieval and reflection, to achieve RE in complex scenarios. Specifically, three major modules are built in AgentRE serving as the tools to help the agent acquire and process various useful information, thereby obtaining improved RE performance. Our extensive experimental results upon two datasets in English and Chinese demonstrate our AgentRE's superior performance, especially in low-resource scenarios. Additionally, the trajectories generated by AgentRE can be refined to construct a high-quality training dataset incorporating different reasoning methods, which can be used to fine-tune smaller models. Code is available at https://github.com/Lightblues/AgentRE.