Numeric Magnitude Comparison Effects in Large Language Models

Raj Shah, Vijay Marupudi, Reba Koenen, Khushi Bhardwaj, Sashank Varma


Abstract
Large Language Models (LLMs) do not differentially represent numbers, which are pervasive in text. In contrast, neuroscience research has identified distinct neural representations for numbers and words. In this work, we investigate how well popular LLMs capture the magnitudes of numbers (e.g., that 4<5) from a behavioral lens. Prior research on the representational capabilities of LLMs evaluates whether they show human-level performance, for instance, high overall accuracy on standard benchmarks. Here, we ask a different question, one inspired by cognitive science: How closely do the number representations of LLMscorrespond to those of human language users, who typically demonstrate the distance, size, and ratio effects? We depend on a linking hypothesis to map the similarities among the model embeddings of number words and digits to human response times. The results reveal surprisingly human-like representations across language models of different architectures, despite the absence of the neural circuitry that directly supports these representations in the human brain. This research shows the utility of understanding LLMs using behavioral benchmarks and points the way to future work on the number of representations of LLMs and their cognitive plausibility.
Anthology ID:
2023.findings-acl.383
Volume:
Findings of the Association for Computational Linguistics: ACL 2023
Month:
July
Year:
2023
Address:
Toronto, Canada
Editors:
Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
6147–6161
Language:
URL:
https://aclanthology.org/2023.findings-acl.383
DOI:
10.18653/v1/2023.findings-acl.383
Bibkey:
Cite (ACL):
Raj Shah, Vijay Marupudi, Reba Koenen, Khushi Bhardwaj, and Sashank Varma. 2023. Numeric Magnitude Comparison Effects in Large Language Models. In Findings of the Association for Computational Linguistics: ACL 2023, pages 6147–6161, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
Numeric Magnitude Comparison Effects in Large Language Models (Shah et al., Findings 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.findings-acl.383.pdf
Video:
 https://aclanthology.org/2023.findings-acl.383.mp4