@article{beyondfirstordertrainingllmswithstochast,
  title = {Beyond First-Order: Training LLMs with Stochastic Conjugate Subgradients and AdamW},
  author = {Di Zhang and Yihang Zhang},
  year = {2025},
  eprint = {2507.01241},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2507.01241},
}