Enhancing Language Models for Financial Relation Extraction with Named Entities and Part-of-Speech

This repository contains the data and codes for the ICLR 2024 Tiny Paper titled "Enhancing Language Models for Financial Relation Extraction with Named Entities and Part-of-Speech" by Menglin Li and Kwan Hui Lim.

Abstract

The Financial Relation Extraction (FinRE) task involves identifying the entities and their relation, given a piece of financial statement/text. To solve this FinRE problem, we propose a simple but effective strategy that improves the performance of pre-trained language models by augmenting them with Named Entity Recognition (NER) and Part-Of-Speech (POS), as well as different approaches to combine these information. Experiments on a financial relations dataset show promising results and highlights the benefits of incorporating NER and POS in existing models. Our dataset and codes are available at https://github.com/kwanhui/FinRelExtract.

Acknowledgement

This work is based on the REFinD dataset and utilizes the publicly available implementations of various models, as referenced in our paper. We thank the authors of these codes and dataset.

Reference

If you find this code useful or use it in your work, please consider citing:

@INPROCEEDINGS { Li-ICLR24,
	AUTHOR = {Menglin Li and Kwan Hui Lim},
	TITLE = {{Enhancing Language Models for Financial Relation Extraction with Named Entities and Part-of-Speech}},
	BOOKTITLE = {{Proceedings of the Twelfth International Conference on Learning Representations (ICLR'24), Tiny Paper Track}},
	MONTH = {May}, 
	YEAR = {2024}
}