Evaluating Spatial Understanding of Large Language Models
Abstract: Large language models (LLMs) show remarkable capabilities across a variety of tasks. Despite the models only seeing text in training, several recent studies suggest that LLM representations implicitly capture aspects of the underlying grounded concepts. Here, we explore LLM representations of a particularly salient kind of grounded knowledge --- spatial relationships. We design natural-language navigation tasks and evaluate the ability of LLMs, in particular GPT-3.5-turbo, GPT-4, and Llama2 series models, to represent and reason about spatial structures. We also compare these abilities to human performance on the same tasks. These tasks reveal substantial variability in LLM performance across different spatial structures, including square, hexagonal, and triangular grids, rings, and trees. In extensive error analysis, we find that LLMs' mistakes reflect both spatial and non-spatial factors. These findings suggest that LLMs appear to capture certain aspects of spatial structure implicitly, but room for improvement remains.
Submission Length: Regular submission (no more than 12 pages of main content)
Code: https://github.com/runopti/SpatialEvalLLM
Assigned Action Editor: ~Yonatan_Bisk1
License: Creative Commons Attribution 4.0 International (CC BY 4.0)
Submission Number: 1679
Loading