Skip to yearly menu bar Skip to main content


Human-Adversarial Visual Question Answering

Sasha Sheng · Amanpreet Singh · Vedanuj Goswami · Jose Magana · Tristan Thrush · Wojciech Galuba · Devi Parikh · Douwe Kiela


Keywords: [ Vision ] [ Adversarial Robustness and Security ]


Performance on the most commonly used Visual Question Answering dataset (VQA v2) is starting to approach human accuracy. However, in interacting with state-of-the-art VQA models, it is clear that the problem is far from being solved. In order to stress test VQA models, we benchmark them against human-adversarial examples. 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. We find that a wide range of state-of-the-art models perform poorly when evaluated on these examples. We conduct an extensive analysis of the collected adversarial examples and provide guidance on future research directions. We hope that this Adversarial VQA (AdVQA) benchmark can help drive progress in the field and advance the state of the art.

Chat is not available.