CV

Education

  • Fall 2026 - Present

    La Jolla, CA, USA

    Ph.D. in Data Science
    University of California San Diego
    Halıcıoğlu Data Science Institute
  • 2022 - 2026

    Shanghai, China

    Bachelor of Science in Computer Science and Technology
    Fudan University
    • GPA: 3.84 / 4.00
  • 2024 Fall

    Austin, TX, USA

    Semester Exchange Student in Computer Science
    The University of Texas at Austin
    • GPA: 4.00 / 4.00
  • 2019 - 2022

    Shanghai, China

    High School
    High School Affiliated to Fudan University

Work Experience

  • Dec. 2024 - Sep. 2025

    Shanghai, China

    Research Intern
    Ant Group NLP Research
    Efficient Language Modeling
    • Co-designed RAMba and implemented sparse retrieval/memory access for efficient long-context modeling (NeurIPS 2025).
    • Analyzed architectural factors behind training-free length generalization in hierarchical sparse attention models (ICLR 2026).

Publications

  1. Understanding and Improving Length Generalization in Hierarchical Sparse Attention Models
    Jiaqi Leng*, Xiang Hu*, Junxiong Wang, Jianguo Li, Wei Wu, and Yucheng Lu
    ICLR 2026
  2. Distilling Token-Trained Models into Byte-Level Models
    Zishuo Bao, Jiaqi Leng, Junxiong Wang, Bowen Peng, and Yucheng Lu
    Preprint, 2026
  3. Hardware-aligned Hierarchical Sparse Attention for Efficient Long-term Memory Access
    Xiang Hu, Jiaqi Leng, Jun Zhao, Kewei Tu, and Wei Wu
    NeurIPS 2025