Combining Quantitative Factors and LLMs for Stock Prediction
New Research Paper: Exploring the Synergy of Quantitative Factors and Newsflow Representations from Large Language Models for Stock Return Prediction.
RAM AI is pleased to share its new research paper, authored by Tian GUO, PhD, Senior Quant, and Emmanuel Hauptmann, CIO & Head of Systematic, which investigates effective model designs and training schemes for combining quantitative factors and newsflow representations derived from Large Language Models (LLMs) in stock return prediction and selection.
Building on our EMNLP 2024 paper, 'Fine-Tuning Large Language Models for Stock Return Prediction Using Newsflow,' this latest work provides empirical and theoretical insights into multimodal fusion learning and mixture modelling with different training strategies.
As the use of multimodal data continues to grow across applications, we hope these studies contribute to the development of more effective multimodal models for quantitative and systematic investing.
Read it on arXiv: https://lnkd.in/exKyzueE
Image Gallery
Legal Disclaimer
Plus d'actualités et d'informations
.png)
.png)
