Human-Adversarial Visual Question Answering

no code implementations • NeurIPS 2021 • Sasha Sheng, Amanpreet Singh, Vedanuj Goswami, Jose Alberto Lopez Magana, Wojciech Galuba, Devi Parikh, Douwe Kiela

Human subjects interact with a state-of-the-art VQA model, and for each image in the dataset, attempt to find a question where the model's predicted answer is incorrect.

Question Answering Visual Question Answering