WHEN FLUE MEETS FLANG: Benchmarks and Large Pre-trained Language Model for Financial Domain
This work addresses the problem of limited evaluation resources and suboptimal models for financial NLP practitioners, offering incremental improvements through domain-specific adaptations.
The authors tackled the lack of specialized benchmarks and pre-trained models for financial NLP by introducing FLUE, a comprehensive evaluation suite, and FLANG, a domain-specific language model that uses financial keywords for masking and novel objectives. Their model outperformed prior methods on various financial NLP tasks, with specific gains reported across benchmarks.
Pre-trained language models have shown impressive performance on a variety of tasks and domains. Previous research on financial language models usually employs a generic training scheme to train standard model architectures, without completely leveraging the richness of the financial data. We propose a novel domain specific Financial LANGuage model (FLANG) which uses financial keywords and phrases for better masking, together with span boundary objective and in-filing objective. Additionally, the evaluation benchmarks in the field have been limited. To this end, we contribute the Financial Language Understanding Evaluation (FLUE), an open-source comprehensive suite of benchmarks for the financial domain. These include new benchmarks across 5 NLP tasks in financial domain as well as common benchmarks used in the previous research. Experiments on these benchmarks suggest that our model outperforms those in prior literature on a variety of NLP tasks. Our models, code and benchmark data are publicly available on Github and Huggingface.