Cross Modality Bias in Visual Question Answering: A Causal View With Possible Worlds VQA
- 21 March 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 26 (15209210) , 8609-8624
Abstract
To increase the generalization capability of VQA systems, many recent studies have tried to de-bias spurious language or vision associations that shortcut the question or image to the answer. Despite these efforts, the literature fails to address the confounding effect of vision and language simultaneously. As a result, when they reduce bias learned from one modality, they usually increase bias from the other. In this paper, we first model a confounding effect that causes language and vision bias simultaneously, then propose a counterfactual inference to remove the influence of this effect. The model trained in this strategy can concurrently and efficiently reduce vision and language bias. To the best of our knowledge, this is the first work to reduce biases resulting from confounding effects of vision and language in VQA, leveraging causal explain-away relations. We accompany our method with an explain-away strategy, pushing the accuracy of the questions with numerical answers results compared to existing methods that have been an open problem. The proposed method outperforms the state-of-the-art methods in VQA-CP v2 datasets.Keywords
All Related Versions
Funding Information
- National Science Foundation (2238208, DGE-1922591, 1909912, DMR-2202124)
- National Nuclear Security Administration (NA0004078)
- NIH (R01EY034562)
- Defense Advanced Research Projects Agency (HR00112220003)
- Center of Excellence in Data Science
- Cisco
This publication has 49 references indexed in Scilit:
- Reducing Language Biases in Visual Question Answering with Visually-Grounded Question Encoder2020
- Learning What Makes a Difference from Counterfactual Examples and Gradient Supervision2020
- Effects of Language on Visual PerceptionTrends in Cognitive Sciences, 2020
- Grounding Visual Explanations2018
- Don't Just Assume; Look and Answer: Overcoming Priors for Visual Question Answering2018
- Human Attention in Visual Question Answering: Do Humans and Deep Networks Look at the Same Regions?Computer Vision and Image Understanding, 2017
- Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answering2017
- VQA: Visual Question AnsweringInternational Journal of Computer Vision, 2016
- Yin and Yang: Balancing and Answering Binary Visual Questions2016
- Stacked Attention Networks for Image Question Answering2016