Detecting ESG topics using domain-specific language models and data augmentation approaches

Abstract: Despite recent advances in deep learning-based language modelling, many NLP tasks in the financial domain remain challenging due to the paucity of appropriately labelled data. Other issues that can limit task performance are differences in word distribution between the general corpora - typically used to pre-train LLMs - and financial corpora, which often exhibit specialized language and symbology. Here, we investigate two approaches that may help to mitigate these issues. Firstly, we experiment with further LLM pre-training using large amounts of in-domain data from business and financial news. We then apply augmentation approaches to increase the size of our dataset for model fine-tuning. We report our findings on an Environmental, Social and Governance (ESG) controversies dataset and demonstrate that both approaches are beneficial to accuracy in classification tasks.