Analyzing the Behavior of Visual Question Answering Models
This work addresses the need for better diagnostic tools in VQA research to guide improvements, though it is incremental as it focuses on analysis rather than new model development.
The paper tackles the problem of understanding the limitations of Visual Question Answering (VQA) models, which typically achieve 60-70% performance, by proposing systematic methods to analyze their behavior, revealing issues like myopia, premature conclusions, and stubbornness.
Recently, a number of deep-learning based models have been proposed for the task of Visual Question Answering (VQA). The performance of most models is clustered around 60-70%. In this paper we propose systematic methods to analyze the behavior of these models as a first step towards recognizing their strengths and weaknesses, and identifying the most fruitful directions for progress. We analyze two models, one each from two major classes of VQA models -- with-attention and without-attention and show the similarities and differences in the behavior of these models. We also analyze the winning entry of the VQA Challenge 2016. Our behavior analysis reveals that despite recent progress, today's VQA models are "myopic" (tend to fail on sufficiently novel instances), often "jump to conclusions" (converge on a predicted answer after 'listening' to just half the question), and are "stubborn" (do not change their answers across images).