Why is Winoground Hard? Investigating Failures in Visuolinguistic Compositionality

Recent visuolinguistic pre-trained models show promising progress on various end tasks such as image retrieval and video captioning. Yet, they fail miserably on the recently proposed Winoground dataset, which challenges models to match paired images and English captions, with items constructed to ov...

Full description

Saved in:
Bibliographic Details
Published inarXiv.org
Main Authors Diwan, Anuj, Berry, Layne, Choi, Eunsol, Harwath, David, Mahowald, Kyle
Format Paper
LanguageEnglish
Published Ithaca Cornell University Library, arXiv.org 03.12.2022
Subjects
Annotations
Datasets
Failure analysis
Grasses
Image resolution
Image retrieval
Inspection
Online AccessGet full text

Cover

More Information
Summary:Recent visuolinguistic pre-trained models show promising progress on various end tasks such as image retrieval and video captioning. Yet, they fail miserably on the recently proposed Winoground dataset, which challenges models to match paired images and English captions, with items constructed to overlap lexically but differ in meaning (e.g., "there is a mug in some grass" vs. "there is some grass in a mug"). By annotating the dataset using new fine-grained tags, we show that solving the Winoground task requires not just compositional language understanding, but a host of other abilities like commonsense reasoning or locating small, out-of-focus objects in low-resolution images. In this paper, we identify the dataset's main challenges through a suite of experiments on related tasks (probing task, image retrieval task), data augmentation, and manual inspection of the dataset. Our analysis suggests that a main challenge in visuolinguistic models may lie in fusing visual and textual representations, rather than in compositional language understanding. We release our annotation and code at https://github.com/ajd12342/why-winoground-hard .
ISSN:2331-8422