Bayesian teaching enables probabilistic reasoning in large language models.

Saved in:
Bibliographic Details
Title: Bayesian teaching enables probabilistic reasoning in large language models.
Authors: Qiu L; Massachusetts Institute of Technology, Cambridge, MA, USA. linluqiu@mit.edu., Sha F; Meta, Menlo Park, CA, USA., Allen K; Google DeepMind, Mountain View, CA, USA.; University of British Columbia, Vancouver, BC, Canada.; Vector Institute, Toronto, ON, Canada., Kim Y; Massachusetts Institute of Technology, Cambridge, MA, USA., Linzen T; Google Research, Mountain View, CA, USA. linzen@google.com.; New York University, New York, NY, USA. linzen@google.com., van Steenkiste S; Google Research, Mountain View, CA, USA. svansteenkiste@google.com.
Source: Nature communications [Nat Commun] 2026 Jan 07; Vol. 17 (1), pp. 1238. Date of Electronic Publication: 2026 Jan 07.
Publication Type: Journal Article
Journal Info: Publisher: Nature Pub. Group Country of Publication: England NLM ID: 101528555 Publication Model: Electronic Cited Medium: Internet ISSN: 2041-1723 (Electronic) Linking ISSN: 20411723 NLM ISO Abbreviation: Nat Commun Subsets: MEDLINE
Database: MEDLINE Ultimate
Description
ISSN:2041-1723
DOI:10.1038/s41467-025-67998-6