Defending Compositionality in Emergent Languages

Michal Auersperger, Pavel Pecina


Abstract
Compositionality has traditionally been understood as a major factor in productivity of language and, more broadly, human cognition. Yet, recently some research started to question its status showing that artificial neural networks are good at generalization even without noticeable compositional behavior. We argue some of these conclusions are too strong and/or incomplete. In the context of a two-agent communication game, we show that compositionality indeed seems essential for successful generalization when the evaluation is done on a suitable dataset.
Anthology ID:
2022.naacl-srw.35
Volume:
Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies: Student Research Workshop
Month:
July
Year:
2022
Address:
Hybrid: Seattle, Washington + Online
Venue:
NAACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
285–291
Language:
URL:
https://aclanthology.org/2022.naacl-srw.35
DOI:
10.18653/v1/2022.naacl-srw.35
Bibkey:
Cite (ACL):
Michal Auersperger and Pavel Pecina. 2022. Defending Compositionality in Emergent Languages. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies: Student Research Workshop, pages 285–291, Hybrid: Seattle, Washington + Online. Association for Computational Linguistics.
Cite (Informal):
Defending Compositionality in Emergent Languages (Auersperger & Pecina, NAACL 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.naacl-srw.35.pdf
Video:
 https://aclanthology.org/2022.naacl-srw.35.mp4
Data
SCAN