Federal Reserve Bank
of San Francisco Federal Reserve Bank
of San Francisco

Philadelphia

  • Can LLMs Credibly Transform the Creation of Panel Data from Diverse Historical Tables?

    Verónica Bäcker-Peral, Vitaly Meursault, Christopher Severen

    Multimodal LLMs offer a watershed change for the digitization of historical tables, enabling low-cost processing centered on domain expertise rather than technical skills. We rigorously validate an LLM-based pipeline on a new panel of historical county-level vehicle registrations. This pipeline is 100 times less expensive than outsourcing, reduces critical parsing errors from 40% to 0.3%, […]

  • Patent Text and Long-Run Innovation Dynamics: The Critical Role of Model Selection

    Ina Ganguli, Jeffrey Lin, Vitaly Meursault, Nicholas Reynolds

    As distorted maps may mislead, Natural Language Processing (NLP) models may misrepresent. How do we know which NLP model to trust? We provide comprehensive guidance for selecting and applying NLP representations of patent text. We develop novel validation tasks to evaluate several leading NLP models. These tasks assess how well candidate models align with both […]

  • Generative AI: A Turning Point for Labor’s Share?

    Lukasz Drozd, Marina M. Tavares

    After years of slow and steady development, generative artificial intelligence (AI) technologies have exploded in popularity, and many experts believe that we are entering a new, AI-driven phase of the Industrial Revolution. The advent of AI as the new engine of growth raises questions about the future of labor. Some have expressed concerns that, in […]

  • PEAD.txt: Post-Earnings-Announcement Drift Using Text

    Vitaly Meursault, Pierre Jinghong Liang, Bryan R. Routledge, and Madeline Marco Scanlon

    Research conducted using AI/ML tools

    We construct a new numerical measure of earnings announcement surprises, standardized unexpected earnings call text (SUE.txt), that does not explicitly incorporate the reported earnings value. SUE.txt generates a text-based post-earnings-announcement drift (PEAD.txt) larger than the classic PEAD. The magnitude of PEAD.txt is considerable even in recent years when the classic PEAD is close to 0. […]

  • Advancing Fairness in Lending Through Machine Learning

    Vitaly Meursault, Daniel Moulton, Larry Santucci, and Nathan Schor, with web adaptation by Kali Aloisi

    Research conducted using AI/ML tools

    Advances in machine learning (ML) provide the opportunity to improve predictions that may expand credit access to more applicants. However, there is concern that gains from advanced models could accrue unequally between demographic groups or do little to reduce existing disparities in credit access. This research explores an approach using ML — paired with setting […]