LLMs Meet Finance: Fine-Tuning Foundation Models for the Open FinLLM Leaderboard

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Rao, Varun, Sun, Youran, Kumar, Mahendra, Mutneja, Tejas, Mukherjee, Agastya, Yang, Haizhao
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917988827922432
author Rao, Varun
Sun, Youran
Kumar, Mahendra
Mutneja, Tejas
Mukherjee, Agastya
Yang, Haizhao
author_facet Rao, Varun
Sun, Youran
Kumar, Mahendra
Mutneja, Tejas
Mukherjee, Agastya
Yang, Haizhao
contents This paper investigates the application of large language models (LLMs) to financial tasks. We fine-tuned foundation models using the Open FinLLM Leaderboard as a benchmark. Building on Qwen2.5 and Deepseek-R1, we employed techniques including supervised fine-tuning (SFT), direct preference optimization (DPO), and reinforcement learning (RL) to enhance their financial capabilities. The fine-tuned models demonstrated substantial performance gains across a wide range of financial tasks. Moreover, we measured the data scaling law in the financial domain. Our work demonstrates the potential of large language models (LLMs) in financial applications.
format Preprint
id arxiv_https___arxiv_org_abs_2504_13125
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LLMs Meet Finance: Fine-Tuning Foundation Models for the Open FinLLM Leaderboard
Rao, Varun
Sun, Youran
Kumar, Mahendra
Mutneja, Tejas
Mukherjee, Agastya
Yang, Haizhao
Computation and Language
Artificial Intelligence
Machine Learning
This paper investigates the application of large language models (LLMs) to financial tasks. We fine-tuned foundation models using the Open FinLLM Leaderboard as a benchmark. Building on Qwen2.5 and Deepseek-R1, we employed techniques including supervised fine-tuning (SFT), direct preference optimization (DPO), and reinforcement learning (RL) to enhance their financial capabilities. The fine-tuned models demonstrated substantial performance gains across a wide range of financial tasks. Moreover, we measured the data scaling law in the financial domain. Our work demonstrates the potential of large language models (LLMs) in financial applications.
title LLMs Meet Finance: Fine-Tuning Foundation Models for the Open FinLLM Leaderboard
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2504.13125