1
POET-X: Memory-efficient LLM Training by Scaling Orthogonal Transformation
ICML 2026 Oral论文，提出通过扩展正交变换实现大模型训练的内存高效方案。
arXiv:2603.05500v2 Announce Type: replace Abstract: Efficient and stable training of large language models (LLMs) remains a core challenge in modern m…