Language Models Represent Space and Time
arXiv:2310.02207
Abstract
The capabilities of large language models (LLMs) have sparked debate over whether such systems just learn an enormous collection of superficial statistics or a set of more coherent and grounded representations that reflect the real world. We find evidence for the latter by analyzing the learned representations of three spatial datasets (world, US, NYC places) and three temporal datasets (historical figures, artworks, news headlines) in the Llama-2 family of models. We discover that LLMs learn linear representations of space and time across multiple scales. These representations are robust to prompting variations and unified across different entity types (e.g. cities and landmarks). In addition, we identify individual "space neurons" and "time neurons" that reliably encode spatial and temporal coordinates. While further investigation is needed, our results suggest modern LLMs learn rich spatiotemporal representations of the real world and possess basic ingredients of a world model.
Cited by in corpus (5)
- Joint Estimation and Prediction of City-wide Delivery Demand: A Large Language Model Empowered Graph-based Learning Approach
- Reward Guidance for Reinforcement Learning Tasks Based on Large Language Models: The LMGT Framework
- Quantifying Geospatial in the Common Crawl Corpus
- Do Large Language Models know who did what to whom?
- Vectorizing string entries for data processing on tables: when are larger language models better?