publications

Research publications and preprints.

2026

  1. Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory
    Hanzuo Liu, Xuan Qi*, Chunyu Liu*, Haotian Zhong*, Yulong Wang, Rayying, Key, Alex Lamb, and Mingyu Gao
    Jul 2026
    arXiv HTML PDF