What Does it Mean for a Neural Network to Learn a "World Model"?
Keywords: Large Language Model, World model
TL;DR: We propose an abstract but precise definition of what it means for a neural net to learn and use a "world model."
Abstract: We propose an abstract but precise definition of what it means for a neural net to learn and use a "world model." The goal is to give an operational meaning to terms that are often used informally, in order to provide a common language for experimental investigation. Our definition is based on ideas from the linear probing literature, and formalizes the notion of a computation that factors through a representation of the data generation process. We also describe a set of conditions to check that such a "world model" is not a trivial consequence of the neural net's data or task.
Primary Area: interpretability and explainable AI
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 10741
Loading