@article{densebackpropagationimprovestrainingfor,
  title = {Dense Backpropagation Improves Training for Sparse Mixture-of-Experts},
  author = {Ashwinee Panda and Vatsal Baherwani and Zain Sarwar and Benjamin Therien and Supriyo Chakraborty and Tom Goldstein},
  year = {2025},
  eprint = {2504.12463},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2504.12463v2},
}