This article is a series of LLM articles, focusing on the translation of "Large Language Models in Finance: A Survey".
A review of research on large language models in finance
Summary
Recent advances in large language models (LLMs) have opened up new possibilities for the application of artificial intelligence in finance. In this article, we provide a practical survey focusing on two key aspects of leveraging LLM for financial tasks: existing solutions and adoption guidelines.
First, we review current approaches to using LLMs in finance, including leveraging pre-trained models with zero or small samples, fine-tuning on domain-specific data, and training custom LLMs from scratch. We summarize the key models and evaluate their performance improvements in financial natural language processing tasks.
Second, we propose a decision-making framework to guide financial professionals in selecting appropriate LLM solutions based on their use case constraints surrounding data, compute, and performance requirements. The framework provides a path from lightweight experimentation to large-scale investment in custom LLM.
Finally, we discuss the limitations and challenges of leveraging LLM in financial applications. Overall, this survey aims to synthesize the state-of-the-art and provide a roadmap for the responsible application of LLM to advance artificial intelligence in finance.
1 Introduction
2 Basics of language model
3 Overview of the application of artificial intelligence in the financial field
4 LLM financial solutions
5 Decision-making process of LLM in financial applications
6 Conclusion
In summary, this paper provides a timely and practical survey of emerging applications of LLM in financial artificial intelligence. We conducted our survey around two key pillars: solutions and adoption guidance.
In the solution, we review various ways to leverage LLMs for financing, including leveraging pre-trained models, fine-tuning on domain data, and training custom LLMs. Experimental results show that in natural language tasks such as sentiment analysis, question answering and summarization, the performance is significantly improved compared with the general LLM.
To provide adoption guidance, we propose a structured framework for selecting the best LLM strategy based on constraints of data availability, computing resources, and performance requirements. The framework aims to balance value and investment by guiding practitioners from low-cost experimentation to rigorous customization.
In summary, this survey synthesizes the latest advances in applying LLM to transform financial AI and provides a practical adoption roadmap. We hope it provides a useful reference for researchers and professionals studying the intersection of LLM and finance. As data sets and computation improve, finance-specific LLM represents an exciting path toward democratizing cutting-edge NLP across the industry.