On July 10, 2026, researchers Bartosz Ziółko and Kacper Dobrzeniewski published a study examining how Large Language Models (LLMs) can enhance fundamental analysis of companies. Their approach utilizes a Retrieval-Augmented Generation (RAG) system to generate investor briefs based on macroeconomic data and SEC filings.
Leveraging Large Language Models in Financial Analysis
The study focuses on the integration of LLMs into financial analysis, specifically how they process data from various reports and documents. By utilizing an API to interface with the gpt-4o model, the researchers aimed to produce comprehensive and automated investor briefs. This method allows for a more dynamic analysis of financial information.
Over a period of four weeks, the team analyzed data from nine companies, providing insights into their performance based on real-time information. The use of LLMs in this context not only streamlines the data processing but also enhances the accuracy of the insights derived from the analysis.
Evaluation of Automated Investor Briefs
To assess the effectiveness of their RAG-based system, the researchers distributed the generated briefs to nine individual investors. The participants evaluated the usefulness of these automated summaries in making informed investment decisions. Feedback from these investors is crucial for refining the approach and ensuring its applicability in real-world scenarios.
