Randomness In Large Language Models: What Researchers Need to Know (And Report)

Guillaume Coqueret, Joan Llull, Florian Oswald, Christophe Pérignon, Christoph Scheuch and Lars Vilhuber
Additional contact information
Guillaume Coqueret: EMLYON Business School
Joan Llull: The Institute for Economic Analysis (IAE); Barcelona School of Economics
Florian Oswald: Università di Torino
Christophe Pérignon: HEC Paris - Finance Department
Christoph Scheuch: Humboldt University of Berlin - Faculty of Economics and Business Administration
Lars Vilhuber: Cornell University - Department of Economics

No 1648, HEC Research Papers Series from HEC Paris

Abstract: Large language models (LLMs) are increasingly used to generate data for research. Typical use cases are classifications, annotations, information extraction, and generation of numerical scores. Unlike conventional measurements, LLM outputs can vary across repeated requests even when the prompt and apparent model settings remain unchanged. This variation arises from deliberate sampling, silent model updates, numerical rounding, or expert routing. Setting a dedicated temperature parameter to zero removes deliberate sampling when that option is available, but it does not eliminate the other sources of randomness. Exact reproduction is therefore generally not possible when using proprietary application programming interfaces. Local execution of open-weight models offers greater control, but reproducibility still depends on the complete hardware and software stack. We illustrate these issues through sentiment classifications of corporate filings and examine their consequences for downstream regression results. We then propose a reporting standard for articles and replication packages, as well as guidance for data editors and authors. Together, these findings and recommendations establish that LLM outputs should be treated as draws from a distribution rather than as fixed measurements.

Keywords: Large Language Models; Reproducibility; Text-as-Data; Measurement Error (search for similar items in EconPapers)
JEL-codes: C18 C63 C81 O33 (search for similar items in EconPapers)
Pages: 19 pages
Date: 2026-07-29
References: Add references at CitEc
Citations:

Downloads: (external link)
https://ssrn.com/abstract=7191580 Full text (text/html)
Our link check indicates that this URL is bad, the error code is: 403 Forbidden (https://ssrn.com/abstract=7191580 [301 Moved Permanently]--> https://www.ssrn.com/abstract=7191580 [302 Found]--> https://papers.ssrn.com/sol3/papers.cfm?abstract_id=7191580)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:ebg:heccah:1648

DOI: 10.2139/ssrn.7191580

Access Statistics for this paper

More papers in HEC Research Papers Series from HEC Paris HEC Paris, 1 Rue de la Libération, 78350 Jouy-en-Josas, France. Contact information at EDIRC.
Bibliographic data for series maintained by David Melon ().