English

Things not Written in Text: Exploring Spatial Commonsense from Visual Signals

Computation and Language 2022-04-28 v2 Computer Vision and Pattern Recognition

Abstract

Spatial commonsense, the knowledge about spatial position and relationship between objects (like the relative size of a lion and a girl, and the position of a boy relative to a bicycle when cycling), is an important part of commonsense knowledge. Although pretrained language models (PLMs) succeed in many NLP tasks, they are shown to be ineffective in spatial commonsense reasoning. Starting from the observation that images are more likely to exhibit spatial commonsense than texts, we explore whether models with visual signals learn more spatial commonsense than text-based PLMs. We propose a spatial commonsense benchmark that focuses on the relative scales of objects, and the positional relationship between people and objects under different actions. We probe PLMs and models with visual signals, including vision-language pretrained models and image synthesis models, on this benchmark, and find that image synthesis models are more capable of learning accurate and consistent spatial knowledge than other models. The spatial knowledge from image synthesis models also helps in natural language understanding tasks that require spatial commonsense.

Keywords

Cite

@article{arxiv.2203.08075,
  title  = {Things not Written in Text: Exploring Spatial Commonsense from Visual Signals},
  author = {Xiao Liu and Da Yin and Yansong Feng and Dongyan Zhao},
  journal= {arXiv preprint arXiv:2203.08075},
  year   = {2022}
}

Comments

Accepted by ACL 2022 main conference

R2 v1 2026-06-24T10:14:24.414Z