IDEAS home Printed from https://ideas.repec.org/p/bdi/opques/qef_935_25.html
   My bibliography  Save this paper

A novel multi-step-prompt approach for LLM-based Q&As on banking supervisory regulations

Author

Listed:
  • Daniele Licari

    (Bank of Italy)

  • Canio Benedetto

    (Bank of Italy)

  • Daniele Bovi

    (Bank of Italy)

  • Praveen Bushipaka

    (Sant'Anna School of Advanced Studies)

  • Alessandro De Gregorio

    (Bank of Italy)

  • Marco De Leonardis

    (Bank of Italy)

  • Tommaso Cucinotta

    (Sant'Anna School of Advanced Studies)

Abstract

This paper investigates the use of large language models (LLMs) in analysing and answering questions related to banking supervisory regulations. We propose a multi-step-prompt approach that enriches the context provided to the LLM with relevant articles from the Capital Requirements Regulation (CRR). We compare our method against standard 'zero-shot' prompting, where the LLM answers are solely based on its pre-trained knowledge, and a standard 'few-shot' prompting, where the LLM is given only a limited number of examples of questions and answers to draw on each time. To assess the quality of the answers returned by the LLM, we also build an 'LLM evaluator' which, for each question, compares the correctness and completeness of the answer resulting from our multi-step prompt approach and from the two standard prompting methods with the official answer made available by the European Banking Authority (EBA), which is taken as a benchmark. Our findings on inquiries concerning Liquidity Risk rules indicate that our multi-step approach significantly improves the quality of LLM-generated answers, offering the analyst a valuable starting point to formulate appropriate answers to particularly complex questions.

Suggested Citation

  • Daniele Licari & Canio Benedetto & Daniele Bovi & Praveen Bushipaka & Alessandro De Gregorio & Marco De Leonardis & Tommaso Cucinotta, 2025. "A novel multi-step-prompt approach for LLM-based Q&As on banking supervisory regulations," Questioni di Economia e Finanza (Occasional Papers) 935, Bank of Italy, Economic Research and International Relations Area.
  • Handle: RePEc:bdi:opques:qef_935_25
    as

    Download full text from publisher

    File URL: https://www.bancaditalia.it/pubblicazioni/qef/2025-0935/QEF_935_25.pdf
    Download Restriction: no
    ---><---

    More about this item

    Keywords

    regulatory Q&A; banking supervisory regulation; Artificial Intelligence; GenAI; GPT-4o; RAG; LLM evaluator;
    All these keywords.

    JEL classification:

    • C45 - Mathematical and Quantitative Methods - - Econometric and Statistical Methods: Special Topics - - - Neural Networks and Related Topics
    • C63 - Mathematical and Quantitative Methods - - Mathematical Methods; Programming Models; Mathematical and Simulation Modeling - - - Computational Techniques
    • G21 - Financial Economics - - Financial Institutions and Services - - - Banks; Other Depository Institutions; Micro Finance Institutions; Mortgages
    • G28 - Financial Economics - - Financial Institutions and Services - - - Government Policy and Regulation

    Statistics

    Access and download statistics

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:bdi:opques:qef_935_25. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    We have no bibliographic references for this item. You can help adding them by using this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: the person in charge (email available below). General contact details of provider: https://edirc.repec.org/data/bdigvit.html .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.