Exploring the Impact of Corpus Diversity on Financial Pretrained Language Models

TL;DR

This paper demonstrates that training financial language models on diverse datasets significantly improves their performance across various financial NLP tasks, outperforming existing models and general PLMs.

Contribution

The study introduces FiLM, a financial language model trained on diverse datasets, showing improved performance over existing financial and general-domain PLMs.

Findings

FiLM outperforms existing financial PLMs.

Diverse training data enhances model generalization.

Improvement observed even on unseen corpus groups.

Abstract

Over the past few years, various domain-specific pretrained language models (PLMs) have been proposed and have outperformed general-domain PLMs in specialized areas such as biomedical, scientific, and clinical domains. In addition, financial PLMs have been studied because of the high economic impact of financial data analysis. However, we found that financial PLMs were not pretrained on sufficiently diverse financial data. This lack of diverse training data leads to a subpar generalization performance, resulting in general-purpose PLMs, including BERT, often outperforming financial PLMs on many downstream tasks. To address this issue, we collected a broad range of financial corpus and trained the Financial Language Model (FiLM) on these diverse datasets. Our experimental results confirm that FiLM outperforms not only existing financial PLMs but also general domain PLMs. Furthermore, we provide empirical evidence that this improvement can be achieved even for unseen corpus groups.