Data-Centric Financial Large Language Models

TL;DR

This paper introduces a data-centric approach for financial large language models (FLLMs) that preprocess and understand data better, using multitask fine-tuning and abductive augmentation reasoning to improve performance on financial tasks.

Contribution

It proposes a novel data-centric methodology with automatic data augmentation for financial LLMs, achieving state-of-the-art results in financial analysis and interpretation.

Findings

FLLM with AAR outperforms baseline models

Achieves state-of-the-art on financial analysis tasks

Open sources a new financial analysis benchmark

Abstract

Large language models (LLMs) show promise for natural language tasks but struggle when applied directly to complex domains like finance. LLMs have difficulty reasoning about and integrating all relevant information. We propose a data-centric approach to enable LLMs to better handle financial tasks. Our key insight is that rather than overloading the LLM with everything at once, it is more effective to preprocess and pre-understand the data. We create a financial LLM (FLLM) using multitask prompt-based finetuning to achieve data pre-processing and pre-understanding. However, labeled data is scarce for each task. To overcome manual annotation costs, we employ abductive augmentation reasoning (AAR) to automatically generate training data by modifying the pseudo labels from FLLM's own outputs. Experiments show our data-centric FLLM with AAR substantially outperforms baseline financial LLMs designed for raw text, achieving state-of-the-art on financial analysis and interpretation tasks. We also open source a new benchmark for financial analysis and interpretation. Our methodology provides a promising path to unlock LLMs' potential for complex real-world domains.