MUTANT: A Training Paradigm for Out-of-Distribution Generalization in Visual Question Answering
Links to Files
Author/Creator
Author/Creator ORCID
Date
Type of Work
Department
Program
Citation of Original Publication
Gokhale, Tejas, Pratyay Banerjee, Chitta Baral, and Yezhou Yang. “MUTANT: A Training Paradigm for Out-of-Distribution Generalization in Visual Question Answering.” Edited by Bonnie Webber, Trevor Cohn, Yulan He, and Yang Liu. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), November 2020, 878–92. https://doi.org/10.18653/v1/2020.emnlp-main.63.
Rights
Attribution 4.0 International
Subjects
Abstract
While progress has been made on the visual question answering leaderboards, models often utilize spurious correlations and priors in datasets under the i.i.d. setting. As such, evaluation on out-of-distribution (OOD) test samples has emerged as a proxy for generalization. In this paper, we present MUTANT, a training paradigm that exposes the model to perceptually similar, yet semantically distinct mutations of the input, to improve OOD generalization, such as the VQA-CP challenge. Under this paradigm, models utilize a consistency-constrained training objective to understand the effect of semantic changes in input (question-image pair) on the output (answer). Unlike existing methods on VQA-CP, MUTANT does not rely on the knowledge about the nature of train and test answer distributions. MUTANT establishes a new state-of-the-art accuracy on VQA-CP with a 10.57% improvement. Our work opens up avenues for the use of semantic input mutations for OOD generalization in question answering.