Authors: Jean Lee, Nicholas Stevens, Soyeon Caren Han, Minseok Song

Published on: February 04, 2024

Impact Score: 8.22

Arxiv code: Arxiv:2402.02315

Summary

• What is new: Provides a comprehensive overview of Financial Large Language Models (FinLLMs), a less explored area compared to general-domain LLMs.
• Why this is important: Limited research and development in FinLLMs despite their potential in transforming financial services.
• What the research proposes: Survey that chronologically overviews the evolution of PLMs to FinLLMs, compares techniques, summarizes performance evaluations, and identifies advanced financial NLP tasks for future research.
• Results: Identifies unique opportunities and challenges for FinLLMs, and compiles essential datasets and benchmarks for AI research in finance on GitHub.

Technical Details

Technological frameworks used: GPT-series, selected open-source LLMs, financial LMs

Models used: Comparative analysis of five techniques across financial PLMs and FinLLMs

Data used: Six benchmark tasks and datasets for performance evaluations; eight advanced financial NLP tasks and datasets

Potential Impact

The financial services sector, including banking, investment, and insurance companies, could be significantly disrupted or benefit from implementing FinLLMs.

Want to implement this idea in a business?

We have generated a startup concept here: FinTuneAI.