I am Longteng Zhang, a final-year PhD candidate in the Thrust of Data Science and Analytics at HKUST(GZ), advised by Xiaowen Chu and Luo Qiong. My research interest lies broadly in optimization and AI topics.

My recent research focuses on building efficient systems capable of reasoning and optimization in long-horizon and complex environments.

I'm always open to discussing research ideas, collaborations, and research or career opportunities. Feel free to reach out via:

WeChat QR code

Publications

Route-Consistent Adaptation for Stable Quantization of Mixture-of-Experts Models with Theoretical Guarantees

Longteng Zhang, S. Wu, Q. Wang, S. Shi, X. Chu

NeurIPS, 2026

SALR: Sparsity-Aware Low-Rank Representation for Efficient Fine-Tuning of Large Language Models

Longteng Zhang, S. Wu, S. Hou, Z. Qing, Z. Zheng, D. Ke, Q. Lin, Q. Wang, S. Shi, X. Chu

AAAI, 2026

Evaluation and Optimization of Gradient Compression for Distributed Deep Learning

L. Zhang, Longteng Zhang, S. Shi, X. Chu, B. Li

ICDCS, 2023

FusionAI: Decentralized Training and Deploying LLMs with Massive Consumer-Level GPUs

Z. Tang, Y. Wang, X. He, Longteng Zhang, X. Pan, Q. Wang, R. Zeng, K. Zhao, S. Shi, B. He, X. Chu

IJCAI, 2023

LoRA-FA: Memory-Efficient Low-Rank Adaptation for Large Language Model Fine-Tuning

Longteng Zhang, L. Zhang, S. Shi, X. Chu, B. Li

Preprint

Dissecting the Runtime Performance of Training, Fine-Tuning, and Inference for Large Language Models

Longteng Zhang, X. Liu, Z. Li, X. Pan, P. Dong, R. Fan, R. Guo, X. Wang, Q. Luo, S. Shi, X. Chu

Preprint

Education

  1. Sep 2023 – Present

    HKUST (Guangzhou)

    Ph.D. Candidate · Data Science and Analytics

    Supervisors: Prof. Xiaowen Chu and Prof. Qiong Luo

  2. Sep 2017 – Jun 2021

    Jilin University

    B.S. · Computer Science and Technology

    Advisor: Prof. Hongliang Li

Experience

  1. May 2026 – Present

    Tencent · Youtu Lab

    Research Intern · Project Up Program

    Efficient post-training for long-horizon agents.

  2. Sep 2022 – Sep 2023

    HKUST (Guangzhou) · DSA Thrust

    Research Assistant

    Efficient LLM training and fine-tuning, runtime benchmarking, and performance analysis of NVIDIA's Hopper architecture.

  3. Sep 2022 – Sep 2023

    Hong Kong Baptist University

    Research Assistant · Department of Computer Science

    Efficient training and optimization of deep learning models.

  4. May 2021 – Oct 2021

    Samsung R&D Institute

    Researcher

    Mobile AI training and inference for Exynos 2200.

Services

Reviewer

  • Conferences: ICML (2025, 2026), NeurIPS (2025, 2026), AAAI (2026), ICLR (2025), IEEE INFOCOM (2025), ACL (2025)
  • Journals: IEEE Network, IEEE Transactions on Network Science and Engineering, IEEE Transactions on Neural Networks and Learning Systems

Teaching Assistant · HKUST(GZ)

  • DSAA5037 · Introduction to Graph Learning (Spring 2025)
  • UCMP6010 · Module C (Fall 2024)