1. Open-FinLLMs: Open Multimodal Large Language Models for Financial Applications
- Author
-
Xie, Qianqian, Li, Dong, Xiao, Mengxi, Jiang, Zihao, Xiang, Ruoyu, Zhang, Xiao, Chen, Zhengyu, He, Yueru, Han, Weiguang, Yang, Yuzhe, Chen, Shunian, Zhang, Yifei, Shen, Lihang, Kim, Daniel, Liu, Zhiwei, Luo, Zheheng, Yu, Yangyang, Cao, Yupeng, Deng, Zhiyang, Yao, Zhiyuan, Li, Haohang, Feng, Duanyu, Dai, Yongfu, Somasundaram, VijayaSai, Lu, Peng, Zhao, Yilun, Long, Yitao, Xiong, Guojun, Smith, Kaleb, Yu, Honghai, Lai, Yanzhao, Peng, Min, Nie, Jianyun, Suchow, Jordan W., Liu, Xiao-Yang, Wang, Benyou, Lopez-Lira, Alejandro, Huang, Jimin, and Ananiadou, Sophia
- Subjects
Computer Science - Computation and Language ,Computer Science - Computational Engineering, Finance, and Science ,Quantitative Finance - Computational Finance - Abstract
Large language models (LLMs) have advanced financial applications, yet they often lack sufficient financial knowledge and struggle with tasks involving multi-modal inputs like tables and time series data. To address these limitations, we introduce \textit{Open-FinLLMs}, a series of Financial LLMs. We begin with FinLLaMA, pre-trained on a 52 billion token financial corpus, incorporating text, tables, and time-series data to embed comprehensive financial knowledge. FinLLaMA is then instruction fine-tuned with 573K financial instructions, resulting in FinLLaMA-instruct, which enhances task performance. Finally, we present FinLLaVA, a multimodal LLM trained with 1.43M image-text instructions to handle complex financial data types. Extensive evaluations demonstrate FinLLaMA's superior performance over LLaMA3-8B, LLaMA3.1-8B, and BloombergGPT in both zero-shot and few-shot settings across 19 and 4 datasets, respectively. FinLLaMA-instruct outperforms GPT-4 and other Financial LLMs on 15 datasets. FinLLaVA excels in understanding tables and charts across 4 multimodal tasks. Additionally, FinLLaMA achieves impressive Sharpe Ratios in trading simulations, highlighting its robust financial application capabilities. We will continually maintain and improve our models and benchmarks to support ongoing innovation in academia and industry., Comment: 33 pages, 13 figures
- Published
- 2024