Author name: Shiju Wang, Yujie Wang, Ao Sun, Fangcheng Fu, Zijian Zhu, Bin Cui, Xu Han, Kaisheng Ma

arXiv:2509.21275v4 Announce Type: replace-cross
Abstract: Long context training is crucial for LLM’s context extension. Existing schemes, such as sequence parallelism, incur substantial communication overhead. Pipeline parallelism (PP) reduces this co…

arXiv:2509.21275v3 Announce Type: replace-cross
Abstract: Long context training is crucial for LLM’s context extension. Existing schemes, such as sequence parallelism, incur substantial communication overhead. Pipeline parallelism (PP) reduces this co…