Syntax-Guided Transformers: Elevating Compositional Generalization and Grounding in Multimodal Environments

Danial Kamali, Parisa Kordjamshidi


Abstract
Compositional generalization, the ability of intelligent models to extrapolate understanding of components to novel compositions, is a fundamental yet challenging facet in AI research, especially within multimodal environments. In this work, we address this challenge by exploiting the syntactic structure of language to boost compositional generalization. This paper elevates the importance of syntactic grounding, particularly through attention masking techniques derived from text input parsing. We introduce and evaluate the merits of using syntactic information in the multimodal grounding problem. Our results on grounded compositional generalization underscore the positive impact of dependency parsing across diverse tasks when utilized with Weight Sharing across the Transformer encoder. The results push the state-of-the-art in multimodal grounding and parameter-efficient modeling and provide insights for future research.
Anthology ID:
2023.genbench-1.10
Volume:
Proceedings of the 1st GenBench Workshop on (Benchmarking) Generalisation in NLP
Month:
December
Year:
2023
Address:
Singapore
Editors:
Dieuwke Hupkes, Verna Dankers, Khuyagbaatar Batsuren, Koustuv Sinha, Amirhossein Kazemnejad, Christos Christodoulopoulos, Ryan Cotterell, Elia Bruni
Venues:
GenBench | WS
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
130–142
Language:
URL:
https://aclanthology.org/2023.genbench-1.10
DOI:
10.18653/v1/2023.genbench-1.10
Bibkey:
Cite (ACL):
Danial Kamali and Parisa Kordjamshidi. 2023. Syntax-Guided Transformers: Elevating Compositional Generalization and Grounding in Multimodal Environments. In Proceedings of the 1st GenBench Workshop on (Benchmarking) Generalisation in NLP, pages 130–142, Singapore. Association for Computational Linguistics.
Cite (Informal):
Syntax-Guided Transformers: Elevating Compositional Generalization and Grounding in Multimodal Environments (Kamali & Kordjamshidi, GenBench-WS 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.genbench-1.10.pdf
Video:
 https://aclanthology.org/2023.genbench-1.10.mp4