Probing GPT-3’s Linguistic Knowledge on Semantic Tasks

Lining Zhang, Mengchen Wang, Liben Chen, Wenxin Zhang


Abstract
GPT-3 has attracted much attention from both academia and industry. However, it is still unclear what GPT-3 has understood or learned especially in linguistic knowledge. Some studies have shown linguistic phenomena including negation and tense are hard to be recognized by language models such as BERT. In this study, we conduct probing tasks focusing on semantic information. Specifically, we investigate GPT-3’s linguistic knowledge on semantic tasks to identify tense, the number of subjects, and the number of objects for a given sentence. We also experiment with different prompt designs and temperatures of the decoding method. Our experiment results suggest that GPT-3 has acquired linguistic knowledge to identify certain semantic information in most cases, but still fails when there are some types of disturbance happening in the sentence. We also perform error analysis to summarize some common types of mistakes that GPT-3 has made when dealing with certain semantic information.
Anthology ID:
2022.blackboxnlp-1.24
Volume:
Proceedings of the Fifth BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP
Month:
December
Year:
2022
Address:
Abu Dhabi, United Arab Emirates (Hybrid)
Editors:
Jasmijn Bastings, Yonatan Belinkov, Yanai Elazar, Dieuwke Hupkes, Naomi Saphra, Sarah Wiegreffe
Venue:
BlackboxNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
297–304
Language:
URL:
https://aclanthology.org/2022.blackboxnlp-1.24
DOI:
10.18653/v1/2022.blackboxnlp-1.24
Bibkey:
Cite (ACL):
Lining Zhang, Mengchen Wang, Liben Chen, and Wenxin Zhang. 2022. Probing GPT-3’s Linguistic Knowledge on Semantic Tasks. In Proceedings of the Fifth BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP, pages 297–304, Abu Dhabi, United Arab Emirates (Hybrid). Association for Computational Linguistics.
Cite (Informal):
Probing GPT-3’s Linguistic Knowledge on Semantic Tasks (Zhang et al., BlackboxNLP 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.blackboxnlp-1.24.pdf