Human-Adversarial Visual Question Answering
Abstract
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.
Cite
Text
Sheng et al. "Human-Adversarial Visual Question Answering." Neural Information Processing Systems, 2021.Markdown
[Sheng et al. "Human-Adversarial Visual Question Answering." Neural Information Processing Systems, 2021.](https://mlanthology.org/neurips/2021/sheng2021neurips-humanadversarial/)BibTeX
@inproceedings{sheng2021neurips-humanadversarial,
title = {{Human-Adversarial Visual Question Answering}},
author = {Sheng, Sasha and Singh, Amanpreet and Goswami, Vedanuj and Magana, Jose and Thrush, Tristan and Galuba, Wojciech and Parikh, Devi and Kiela, Douwe},
booktitle = {Neural Information Processing Systems},
year = {2021},
url = {https://mlanthology.org/neurips/2021/sheng2021neurips-humanadversarial/}
}