Reproduce “Layerwise Repacking: Variable-Length Hybrid-Attention Training without an Extra Redistribution Pass” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Jiahao Zhang, Junwei Feng, Dongliang Huang, Xiaoling Zhou, Yiqiao Liao, Chengru Song
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.