@article{beyondsunkcostsboostingllmpretrainingeff,
  title = {Beyond Sunk Costs: Boosting LLM Pre-training Efficiency via Orthogonal Growth of Mixture-of-Experts},
  author = {Ruizhe Wang and Yucheng Ding and Xiao Liu and Yaoxiang Wang and Peng Cheng and Baining Guo and Zhengjun Zha and Yeyun Gong},
  year = {2025},
  eprint = {2510.08008},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2510.08008},
}