A Large-Scale Multilingual Study of Visual Constraints on Linguistic Selection of Descriptions

Uri Berger, Lea Frermann, Gabriel Stanovsky, Omri Abend


Abstract
We present a large, multilingual study into how vision constrains linguistic choice, covering four languages and five linguistic properties, such as verb transitivity or use of numerals. We propose a novel method that leverages existing corpora of images with captions written by native speakers, and apply it to nine corpora, comprising 600k images and 3M captions. We study the relation between visual input and linguistic choices by training classifiers to predict the probability of expressing a property from raw images, and find evidence supporting the claim that linguistic properties are constrained by visual context across languages. We complement this investigation with a corpus study, taking the test case of numerals. Specifically, we use existing annotations (number or type of objects) to investigate the effect of different visual conditions on the use of numeral expressions in captions, and show that similar patterns emerge across languages. Our methods and findings both confirm and extend existing research in the cognitive literature. We additionally discuss possible applications for language generation.
Anthology ID:
2023.findings-eacl.172
Volume:
Findings of the Association for Computational Linguistics: EACL 2023
Month:
May
Year:
2023
Address:
Dubrovnik, Croatia
Editors:
Andreas Vlachos, Isabelle Augenstein
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
2285–2299
Language:
URL:
https://aclanthology.org/2023.findings-eacl.172
DOI:
10.18653/v1/2023.findings-eacl.172
Bibkey:
Cite (ACL):
Uri Berger, Lea Frermann, Gabriel Stanovsky, and Omri Abend. 2023. A Large-Scale Multilingual Study of Visual Constraints on Linguistic Selection of Descriptions. In Findings of the Association for Computational Linguistics: EACL 2023, pages 2285–2299, Dubrovnik, Croatia. Association for Computational Linguistics.
Cite (Informal):
A Large-Scale Multilingual Study of Visual Constraints on Linguistic Selection of Descriptions (Berger et al., Findings 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.findings-eacl.172.pdf
Video:
 https://aclanthology.org/2023.findings-eacl.172.mp4