@article{nonuniformtensorparallelismmitigatinggpu,
  title = {Nonuniform-Tensor-Parallelism: Mitigating GPU failure impact for Scaled-up LLM Training},
  author = {Daiyaan Arfeen and Dheevatsa Mudigere and Ankit More and Bhargava Gopireddy and Ahmet Inci and Gregory R. Ganger},
  year = {2025},
  eprint = {2504.06095},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2504.06095v1},
}