Can large language models create new knowledge for spatial reasoning tasks?

Greatrix, Thomas, Whitaker, Roger ORCID: https://orcid.org/0000-0002-8473-1913, Turner, Liam ORCID: https://orcid.org/0000-0003-4877-5289 and Colombo, Walter 2024. Can large language models create new knowledge for spatial reasoning tasks? Cornell University.

PDF - Submitted Pre-Print Version Download (1MB)

Abstract

The potential for Large Language Models (LLMs) to generate new information offers a potential step change for research and innovation. This is challenging to assert as it can be difficult to determine what an LLM has previously seen during training, making "newness" difficult to substantiate. In this paper we observe that LLMs are able to perform sophisticated reasoning on problems with a spatial dimension, that they are unlikely to have previously directly encountered. While not perfect, this points to a significant level of understanding that state-of-the-art LLMs can now achieve, supporting the proposition that LLMs are able to yield significant emergent properties. In particular, Claude 3 is found to perform well in this regard.

Item Type:	Working paper
Date Type:	Submission
Status:	Submitted
Schools:	Schools > Computer Science & Informatics
Publisher:	Cornell University
Last Modified:	09 Oct 2026 14:09
URI:	https://orca.cardiff.ac.uk/id/eprint/170237

Actions (repository staff only)

Edit Item