Effectiveness of Large-Language Models in Recognizing Spatially Intensive Statistical Data
Published in Singapore Institute of Technology, 2024
A statistical quantity associated with a geographic enumeration unit is termed “intensive” if its value is, at least approximately, independent of the unit’s spatial extent. This study evaluates the ability of large language models (LLMs) to identify whether a quantity is intensive. Overall, certain combinations of LLMs, intensiveness definitions, and data descriptions performed well in the classification. [pdf]
Recommended citation: Tharatipyakul, Atima, Haw Yuh Loh, Simon T. Perrault, Yong Wang, and Michael Thorsten Gastner. "Effectiveness of Large-Language Models in Recognizing Spatially Intensive Statistical Data." Singapore Institute of Technology, 2024.
Download Paper