VQA-LOL: Visual Question Answering Under the Lens of Logic

Tejas Gokhale, Pratyay Banerjee, Chitta Baral, Yezhou Yang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations

Abstract

Logical connectives and their implications on the meaning of a natural language sentence are a fundamental aspect of understanding. In this paper, we investigate whether visual question answering (VQA) systems trained to answer a question about an image, are able to answer the logical composition of multiple such questions. When put under this Lens of Logic, state-of-the-art VQA models have difficulty in correctly answering these logically composed questions. We construct an augmentation of the VQA dataset as a benchmark, with questions containing logical compositions and linguistic transformations (negation, disjunction, conjunction, and antonyms). We propose our Lens of Logic (LOL) model which uses question-attention and logic-attention to understand logical connectives in the question, and a novel Fréchet-Compatibility Loss, which ensures that the answers of the component questions and the composed question are consistent with the inferred logical operation. Our model shows substantial improvement in learning logical compositions while retaining performance on VQA. We suggest this work as a move towards robustness by embedding logical connectives in visual understanding.

Original languageEnglish (US)
Title of host publicationComputer Vision – ECCV 2020 - 16th European Conference 2020, Proceedings
EditorsAndrea Vedaldi, Horst Bischof, Thomas Brox, Jan-Michael Frahm
PublisherSpringer Science and Business Media Deutschland GmbH
Pages379-396
Number of pages18
ISBN (Print)9783030585884
DOIs
StatePublished - 2020
Event16th European Conference on Computer Vision, ECCV 2020 - Glasgow, United Kingdom
Duration: Aug 23 2020Aug 28 2020

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume12366 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference16th European Conference on Computer Vision, ECCV 2020
Country/TerritoryUnited Kingdom
CityGlasgow
Period8/23/208/28/20

Keywords

  • Logical robustness
  • Visual question answering

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint

Dive into the research topics of 'VQA-LOL: Visual Question Answering Under the Lens of Logic'. Together they form a unique fingerprint.

Cite this