Please use this identifier to cite or link to this item:
https://hdl.handle.net/2440/136027
Citations | ||
Scopus | Web of Science® | Altmetric |
---|---|---|
?
|
?
|
Type: | Conference paper |
Title: | Human-like property induction is a challenge for large language models |
Author: | Han, S.J. Ransom, K.J. Perfors, A. Kemp, C. |
Citation: | Proceedings of the 44th Annual Conference of the Cognitive Science Society (CogSci 2022), 2022, pp.2782-2788 |
Publisher: | Cognitive Science Society, University of California |
Issue Date: | 2022 |
Conference Name: | Annual Conference of the Cognitive Science Society (CogSci) (27 Jul 2022 - 30 Jul 2022 : Toronto, Canada and Virtual Online) |
Statement of Responsibility: | Simon Jerome Han, Keith J. Ransom, Andrew Perfors, Charles Kemp |
Abstract: | The impressive recent performance of large language models such as GPT-3 has led many to wonder to what extent they can serve as models of general intelligence or are similar to human cognition. We address this issue by applying GPT-3 to a classic problem in human inductive reasoning known as property induction. Our results suggest that while GPT-3 can qualitatively mimic human performance for some inductive phenomena (especially those that depend primarily on similarity relationships), it reasons in a qualitatively distinct way on phenomena that require more theoretical understanding. We propose that this emerges due to the reasoning abilities of GPT-3 rather than its underlying representations, and suggest that increasing its scale is unlikely to change this pattern. |
Keywords: | reasoning; property induction; neural networks; GPT-3; AI |
Rights: | ©2022 The Author(s). This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY). |
DOI: | 10.31234/osf.io/6mkjy |
Published version: | https://escholarship.org/uc/item/3w84q1s1 |
Appears in Collections: | Psychology publications |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
hdl_136027.pdf | Published version | 3.14 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.