Check It Again: Progressive Visual Question Answering via Visual Entailment
This addresses the issue of superficial correlations in VQA models, offering a domain-specific solution for more reliable question answering.
The paper tackles the problem of language priors in Visual Question Answering by proposing a select-and-rerank framework based on visual entailment, which verifies candidate answers against the image, achieving a 7.55% improvement in accuracy on VQA-CP v2.
While sophisticated Visual Question Answering models have achieved remarkable success, they tend to answer questions only according to superficial correlations between question and answer. Several recent approaches have been developed to address this language priors problem. However, most of them predict the correct answer according to one best output without checking the authenticity of answers. Besides, they only explore the interaction between image and question, ignoring the semantics of candidate answers. In this paper, we propose a select-and-rerank (SAR) progressive framework based on Visual Entailment. Specifically, we first select the candidate answers relevant to the question or the image, then we rerank the candidate answers by a visual entailment task, which verifies whether the image semantically entails the synthetic statement of the question and each candidate answer. Experimental results show the effectiveness of our proposed framework, which establishes a new state-of-the-art accuracy on VQA-CP v2 with a 7.55% improvement.