SusGen-GPT: A Data-Centric LLM for Financial NLP and Sustainability Report Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wu, Qilong, Xiang, Xiaoneng, Huang, Hejia, Wang, Xuan, Jie, Yeo Wei, Satapathy, Ranjan, Filho, Ricardo Shirota, Veeravalli, Bharadwaj
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913613260783616
author Wu, Qilong
Xiang, Xiaoneng
Huang, Hejia
Wang, Xuan
Jie, Yeo Wei
Satapathy, Ranjan
Filho, Ricardo Shirota
Veeravalli, Bharadwaj
author_facet Wu, Qilong
Xiang, Xiaoneng
Huang, Hejia
Wang, Xuan
Jie, Yeo Wei
Satapathy, Ranjan
Filho, Ricardo Shirota
Veeravalli, Bharadwaj
contents The rapid growth of the financial sector and the rising focus on Environmental, Social, and Governance (ESG) considerations highlight the need for advanced NLP tools. However, open-source LLMs proficient in both finance and ESG domains remain scarce. To address this gap, we introduce SusGen-30K, a category-balanced dataset comprising seven financial NLP tasks and ESG report generation, and propose TCFD-Bench, a benchmark for evaluating sustainability report generation. Leveraging this dataset, we developed SusGen-GPT, a suite of models achieving state-of-the-art performance across six adapted and two off-the-shelf tasks, trailing GPT-4 by only 2% despite using 7-8B parameters compared to GPT-4's 1,700B. Based on this, we propose the SusGen system, integrated with Retrieval-Augmented Generation (RAG), to assist in sustainability report generation. This work demonstrates the efficiency of our approach, advancing research in finance and ESG.
format Preprint
id arxiv_https___arxiv_org_abs_2412_10906
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle SusGen-GPT: A Data-Centric LLM for Financial NLP and Sustainability Report Generation
Wu, Qilong
Xiang, Xiaoneng
Huang, Hejia
Wang, Xuan
Jie, Yeo Wei
Satapathy, Ranjan
Filho, Ricardo Shirota
Veeravalli, Bharadwaj
Computation and Language
Computational Engineering, Finance, and Science
Machine Learning
Computational Finance
The rapid growth of the financial sector and the rising focus on Environmental, Social, and Governance (ESG) considerations highlight the need for advanced NLP tools. However, open-source LLMs proficient in both finance and ESG domains remain scarce. To address this gap, we introduce SusGen-30K, a category-balanced dataset comprising seven financial NLP tasks and ESG report generation, and propose TCFD-Bench, a benchmark for evaluating sustainability report generation. Leveraging this dataset, we developed SusGen-GPT, a suite of models achieving state-of-the-art performance across six adapted and two off-the-shelf tasks, trailing GPT-4 by only 2% despite using 7-8B parameters compared to GPT-4's 1,700B. Based on this, we propose the SusGen system, integrated with Retrieval-Augmented Generation (RAG), to assist in sustainability report generation. This work demonstrates the efficiency of our approach, advancing research in finance and ESG.
title SusGen-GPT: A Data-Centric LLM for Financial NLP and Sustainability Report Generation
topic Computation and Language
Computational Engineering, Finance, and Science
Machine Learning
Computational Finance
url https://arxiv.org/abs/2412.10906