New system fuses LLMs and financial data for Fed policy stance classification

Researchers have developed a novel system called LabelFusion-TS that integrates large language models (LLMs), transformer encoders, and financial time series data to classify monetary policy stances from Federal Reserve communications. This approach aims to improve classification accuracy by incorporating market context beyond just text. The system demonstrated superior performance, achieving a 70.2% weighted F1 score, outperforming a zero-shot LLM by a significant margin, especially with limited labeled data. AI

IMPACT This research suggests that incorporating financial time series data can significantly enhance the accuracy of LLM-based text classification in specialized domains like monetary policy analysis.

RANK_REASON The cluster contains an academic paper detailing a new method for text classification using LLMs and financial data. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• Federal Open Market Committee
• Federal Reserve System
• Financial time series prediction using least squares support vector machines within the evidence framework
• LabelFusion-TS
• large-language models
• Roberta
• transformer encoders

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →