Emergent Analogical Reasoning in Large Language Models

arxiv:2212.09196

Original source · Paper JSON

Reasoning and the "physics" of language models · 2022-12-19

Cites (2) · Cited by (2)

Complete cleaned Markdown · Source Markdown including bibliography

Read in parts

1. Part 1
2. Part 2
3. Part 3
4. Part 4
5. Part 5

Contents

• Emergent Analogical Reasoning in Large Language Models 0 0 footnotetext: Published at Nature Human Behaviour (2023) https://doi.org/10.1038/s41562-023-01659-w
• Abstract
• 1 Introduction
• 2 Results
• 2.1 Matrix reasoning problems
• 2.2 Letter string analogies
• 2.3 Four-term verbal analogies
• 2.4 Story analogies
• 2.5 Analogical problem-solving
• 3 Discussion
• 4 Methods
• 4.1 Code
• 4.2 GPT-3
• 4.3 Digit Matrices
• 4.3.1 Dataset
• 4.3.2 Human behavioral experiments
• 4.3.3 Evaluating GPT-3
• 4.3.4 Statistical analyses
• 4.4 Letter string analogies
• 4.4.1 Problem set
• 4.4.2 Evaluating GPT-3
• 4.4.3 Human behavioral experiment
• 4.4.4 Statistical analyses
• 4.5 Four-term verbal analogies
• 4.6 Story analogies
• 4.6.1 Materials
• 4.6.2 Human behavioral experiment
• 4.6.3 Evaluating GPT-3
• 4.6.4 Evaluating GPT-4
• 4.6.5 Statistical analyses
• 4.7 Analogical problem-solving
• Data Availability
• Code Availability
• Acknowledgements
• Author Contributions Statement
• Competing Interests Statement