Does GPT-3 Grasp Metaphors? Identifying Metaphor Mappings with Generative Language Models

Lennart Wachowiak, Dagmar Gromann


Abstract
Conceptual metaphors present a powerful cognitive vehicle to transfer knowledge structures from a source to a target domain. Prior neural approaches focus on detecting whether natural language sequences are metaphoric or literal. We believe that to truly probe metaphoric knowledge in pre-trained language models, their capability to detect this transfer should be investigated. To this end, this paper proposes to probe the ability of GPT-3 to detect metaphoric language and predict the metaphor’s source domain without any pre-set domains. We experiment with different training sample configurations for fine-tuning and few-shot prompting on two distinct datasets. When provided 12 few-shot samples in the prompt, GPT-3 generates the correct source domain for a new sample with an accuracy of 65.15% in English and 34.65% in Spanish. GPT’s most common error is a hallucinated source domain for which no indicator is present in the sentence. Other common errors include identifying a sequence as literal even though a metaphor is present and predicting the wrong source domain based on specific words in the sequence that are not metaphorically related to the target domain.
Anthology ID:
2023.acl-long.58
Volume:
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Month:
July
Year:
2023
Address:
Toronto, Canada
Editors:
Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1018–1032
Language:
URL:
https://aclanthology.org/2023.acl-long.58
DOI:
10.18653/v1/2023.acl-long.58
Bibkey:
Cite (ACL):
Lennart Wachowiak and Dagmar Gromann. 2023. Does GPT-3 Grasp Metaphors? Identifying Metaphor Mappings with Generative Language Models. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1018–1032, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
Does GPT-3 Grasp Metaphors? Identifying Metaphor Mappings with Generative Language Models (Wachowiak & Gromann, ACL 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.acl-long.58.pdf
Video:
 https://aclanthology.org/2023.acl-long.58.mp4