Improving the trustworthiness of image classification models by utilizing bounding box annotations. NeurIPS workshop on Interpretable Inductive Biases and Physically Structured Learning [2020].

If you use this code please cite:

@article{kc2021improving,
  title={Improving the trustworthiness of image classification models by utilizing bounding-box annotations},
  author={KC, Dharma and Zhang, Chicheng},
  journal={arXiv preprint arXiv:2108.10131},
  year={2021}
}