EUVIMEDEUVIMED entdecken
ÜbersichtsarbeitOpen Access · CC BY 4.0

Levels of analysis for large language models

Alexander Ku, Declan Campbell, Xuechunzi Bai, Jiayi Geng, Ryan Liu, Raja Marjieh, R Thomas McCoy, Andrew Nam, Ilia Sucholutsky, Liyi Zhang, Jian-Qiao Zhu, Thomas Griffiths iD

Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences · 14.05.2026 · 384 (2320)

DOI: 10.1098/rsta.2025.0012 ↗

Kurz gefasst

Übersichtsarbeit in der Zeitschrift Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, online erschienen am 14.05.2026. Verfasst von Alexander Ku und 11 weiteren Autor:innen. Beteiligte Einrichtung: Princeton University , Princeton, NJ, USA. Hauptthemen laut MEDLINE: Artificial Intelligence, Cognitive Science. Die Arbeit nennt 140 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NOMIS Stiftung.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Modern artificial intelligence systems, such as large language models, are increasingly powerful but also increasingly hard to understand. Recognizing this problem as analogous to the historical difficulties in understanding the human mind, we argue that methods developed in cognitive science can be useful for understanding large language models. We propose a framework for applying these methods based on the levels of analysis that David Marr proposed for studying information processing systems. By revisiting established cognitive science techniques relevant to each level and illustrating their potential to yield insights into the behaviour and internal organization of large language models, we aim to provide a toolkit for making sense of these new kinds of minds. This article is part of the theme issue ‘World models in natural and artificial intelligence’.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Artificial IntelligenceCognitive ScienceLarge Language ModelsHumansGenerative Artificial Intelligence

Stichwörter der Autor:innen

cognitive sciencelevels of analysismachine behaviour

Autor:innen und Institutionen

  • Princeton University , Princeton, NJ, USA.
  • The University of Chicago , Chicago, IL, USA.
  • Yale University , New Haven, CT, USA.
  • New York University , New York, NY, USA.

ORCID-Profile: Thomas Griffiths

Förderung

  • NOMIS Stiftung

Zitieren

APA

Ku, A., Campbell, D., Bai, X., Geng, J., Liu, R., Marjieh, R., McCoy, R. T., Nam, A., Sucholutsky, I., Zhang, L., Zhu, J. Q., & Griffiths, T. (2026). Levels of analysis for large language models. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 384(2320). https://doi.org/10.1098/rsta.2025.0012

Vancouver

Ku A, Campbell D, Bai X, Geng J, Liu R, Marjieh R, et al. Levels of analysis for large language models. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences. 2026;384(2320). doi:10.1098/rsta.2025.0012

BibTeX
@article{Ku2026,
  author = {Ku, Alexander and Campbell, Declan and Bai, Xuechunzi and Geng, Jiayi and Liu, Ryan and Marjieh, Raja and McCoy, R Thomas and Nam, Andrew and Sucholutsky, Ilia and Zhang, Liyi and Zhu, Jian-Qiao and Griffiths, Thomas},
  title = {Levels of analysis for large language models},
  journal = {Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences},
  year = {2026},
  volume = {384},
  number = {2320},
  doi = {10.1098/rsta.2025.0012}
}
Kostenlos starten →