Self-Critical Reasoning for Robust Visual Question Answering

Abstract

Visual Question Answering (VQA) deep-learning systems tend to capture superficial statistical correlations in the training data because of strong language priors and fail to generalize to test data with a significantly different question-answer (QA) distribution. To address this issue, we introduce a self-critical training objective that ensures that visual explanations of correct answers match the most influential image regions more than other competitive answer candidates. The influential regions are either determined from human visual/textual explanations or automatically from just significant words in the question and answer. We evaluate our approach on the VQA generalization task using the VQA-CP dataset, achieving a new state-of-the-art i.e., 49.5% using textual explanations and 48.5% using automatically annotated regions.

Jialin Wu

Research Scientist

I am interested in enhancing the capabilities of image generation models on info-seeking (world knowledge) queries. Some research questions I am exploring include (1) utilizing search signals during the pre/post-training phases as well as during inference for image generation, and (2) enhancing the factual accuracy of images produced in response to info-seeking queries.