% Key papers for Batch normalization improvements (scime atlas, https://atlas.scime.ai). Authors are not included: add them from the paper pages.

@inproceedings{neurips2017_batch_56172,
  title = {{Batch Renormalization: Towards Reducing Minibatch Dependence in Batch-Normalized Models}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2017},
  url = {https://nips.cc/virtual/2017/poster/8983},
}

@inproceedings{cvpr2018_in_13221,
  title = {{In-Place Activated BatchNorm for Memory-Optimized Training of DNNs}},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year = {2018},
  url = {https://openaccess.thecvf.com/content_cvpr_2018/html/Bulo_In-Place_Activated_BatchNorm_CVPR_2018_paper.html},
}

@inproceedings{cvpr2019_understanding_14959,
  title = {{Understanding the Disharmony Between Dropout and Batch Normalization by Variance Shift}},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year = {2019},
}

@inproceedings{icml2021_high_39423,
  title = {{High-Performance Large-Scale Image Recognition Without Normalization}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2021},
  url = {https://icml.cc/virtual/2021/poster/9957},
}

@inproceedings{neurips2018_how_57205,
  title = {{How Does Batch Normalization Help Optimization?}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2018},
  url = {https://nips.cc/virtual/2018/poster/11257},
}

@inproceedings{cvpr2018_decorrelated_12951,
  title = {{Decorrelated Batch Normalization}},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year = {2018},
  url = {https://openaccess.thecvf.com/content_cvpr_2018/html/Huang_Decorrelated_Batch_Normalization_CVPR_2018_paper.html},
}

@inproceedings{cvpr2019_iterative_14339,
  title = {{Iterative Normalization: Beyond Standardization Towards Efficient Whitening}},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year = {2019},
}

@inproceedings{cvpr2020_filter_15550,
  title = {{Filter Response Normalization Layer: Eliminating Batch Dependence in the Training of Deep Neural Networks}},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year = {2020},
}

@inproceedings{cvpr2021_cross_16766,
  title = {{Cross-Iteration Batch Normalization}},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year = {2021},
}

@inproceedings{neurips2018_norm_57448,
  title = {{Norm matters: efficient and accurate normalization schemes in deep networks}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2018},
  url = {https://nips.cc/virtual/2018/poster/11227},
}

@inproceedings{icml2016_normalization_35923,
  title = {{Normalization Propagation: A Parametric Technique for Removing Internal Covariate Shift in Deep Networks}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2016},
  url = {https://proceedings.mlr.press/v48/arpitb16.html},
}

@inproceedings{neurips2020_batch_59430,
  title = {{Batch Normalization Biases Residual Blocks Towards the Identity Function in Deep Networks}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2020},
  url = {https://nips.cc/virtual/2020/poster/18721},
}

@inproceedings{iclr2017_normalizing_28373,
  title = {{Normalizing the Normalizers: Comparing and Extending Network Normalization Schemes}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2017},
  url = {https://openreview.net/forum?id=rk5upnsxe},
}

@inproceedings{neurips2017_riemannian_56630,
  title = {{Riemannian approach to batch normalization}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2017},
  url = {https://nips.cc/virtual/2017/poster/9298},
}

@inproceedings{iclr2017_recurrent_28347,
  title = {{Recurrent Batch Normalization}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2017},
  url = {https://openreview.net/forum?id=r1VdcHcxx},
}

@inproceedings{iclr2019_towards_28912,
  title = {{Towards Understanding Regularization in Batch Normalization}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2019},
  url = {https://iclr.cc/virtual/2019/poster/835},
}

@inproceedings{iclr2020_four_29478,
  title = {{Four Things Everyone Should Know to Improve Batch Normalization}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2020},
  url = {https://openreview.net/forum?id=HJx8HANFDH},
}

@inproceedings{iclr2020_an_29785,
  title = {{An Exponential Learning Rate Schedule for Deep Learning}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2020},
  url = {https://openreview.net/forum?id=rJg8TeSFDH},
}

@inproceedings{aaai2021_training_1550,
  title = {{Training Binary Neural Network without Batch Normalization for Image Super-Resolution}},
  booktitle = {AAAI Conference on Artificial Intelligence (AAAI)},
  year = {2021},
  url = {https://aaai.org/papers/01700-16263-training-binary-neural-network-without-batch-normalization-for-image-super-resolution/},
}

@inproceedings{aaai2020_new_102184,
  title = {{New Interpretations of Normalization Methods in Deep Learning}},
  booktitle = {AAAI Conference on Artificial Intelligence (AAAI)},
  year = {2020},
  url = {https://ojs.aaai.org/index.php/AAAI/article/view/6046},
}

@inproceedings{aaai2024_batch_5272,
  title = {{Batch Normalization Is Blind to the First and Second Derivatives of the Loss}},
  booktitle = {AAAI Conference on Artificial Intelligence (AAAI)},
  year = {2024},
  url = {https://ojs.aaai.org/index.php/AAAI/article/view/29978},
}

@inproceedings{icml2025_merit_47089,
  title = {{MERIT: Maximum-normalized Element-wise Ratio for Language Model Large-batch Training}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2025},
  url = {https://icml.cc/virtual/2025/poster/45497},
}

@inproceedings{neurips2025_critical_77781,
  title = {{Critical Batch Size Revisited: A Simple Empirical Approach to Large-Batch Language Model Training}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=XUKUx7Xu89},
}

@inproceedings{neurips2026_the_93078,
  title = {{The BatchNorm Illusion: Diagnosing Normalization Artifacts in Machine Unlearning Evaluation}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2026},
  url = {https://neurips.cc/virtual/2026/poster/155403},
}

@inproceedings{iclr2024_batch_35577,
  title = {{Batch normalization is sufficient for universal function approximation in CNNs}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2024},
  url = {https://iclr.cc/virtual/2024/poster/17496},
}