Zicheng Zhang 张紫程

Biography

I am currently a Senior Researcher at JINGDONG (JD.com) and a member of its Tech Genius Team (TGT) program.

My recent research focuses on large language model training and applications, particularly efficient adaptation, data selection, and reasoning, across JD's retail ecosystem and a product corpus at tens-of-billions scale.

I received my Ph.D. in Operations Research in 2024 from the School of Mathematical Sciences, University of Chinese Academy of Sciences, advised by Prof. Congying Han and Prof. Tiande Guo. Before that, I earned a B.Econ. in Financial Mathematics in 2019 from Zhengzhou University.

Research Interests

My research spans large language model training, data-centric learning, generative models, multimodal intelligence, reinforcement learning, and 3D vision. I am particularly interested in methods that connect theoretically grounded design with practical systems at scale.

• LLM training, efficient adaptation, and data selection
• Generative models for image, video, and 3D content
• Machine learning and optimization
• Large-scale retail intelligence and applications

Experience

• JINGDONG, Senior Researcher, Tech Genius Team (TGT) Current
  Large language model training and applications across JD's core retail ecosystem.

• Ant Group, Research Intern, Digital Human Jun 2023 – Dec 2023
  Developed high-fidelity NeRF and tetrahedra-based digital human systems for Ant Digital Human Platform.

• Meitu Imaging & Vision Lab, Research Intern, Generative Vision Mar 2023 – Jun 2023
  Explored diffusion-based portrait stylization and video generation for research and applications in Meitu APP.

• JD Explore Academy, Research Intern, Digital Human Jun 2021 – Feb 2023
  Developed visual algorithms for Yanxi digital humans, spanning model design, training and optimization.

Selected Honors and Awards

• NeurIPS 2026 Top Reviewer

• JD Retail Star Employee Award (2026)

• Beijing Yicheng Elite Talent Award (2025)

• CAS President Prize/Scholarship (2024)

• National Scholarship for Doctoral Students (2023)

• CAS AMSS President Prize/Scholarship (2022)

Selected Publications and Preprints

Large Language Model Training

• The Hidden Power of Scaling Factor in LoRA Optimization
  Zicheng Zhang, Haoran Li, Jiaxing Wang, Guoqiang Gong, Anqi Li, Yudong Hu, et al.
  NeurIPS 2026.

• BLADE: Scalable Bi-level Adaptive Data Selection for LLM Training
  Jiaxing Wang, Deping Xiang, Jin Xu, Zirui Liu, Zicheng Zhang, Guoqiang Gong, et al.
  NeurIPS 2026.

• Top-Down Compression: Revisit Efficient Vision Token Projection for Visual Instruction Tuning
  Bonan Li*, Zicheng Zhang*, Songhua Liu, Weihao Yu, Xinchao Wang.
  IEEE TIP 2026.

• JoyAgents-R1: Joint Evolution Dynamics for Versatile Multi-LLM Agents with Reinforcement Learning
  Ai Han, Junxing Hu, Pu Wei, Zhiqian Zhang, Yuhang Guo, Jiawei Lu, Zicheng Zhang.
  Preprint, 2025.

• The Primacy of Magnitude in Low-Rank Adaptation
  Zicheng Zhang, Haoran Li, Yifeng Zhang, Guoqiang Gong, Jiaxing Wang, Junxing Hu, et al.
  NeurIPS 2025 Spotlight.

• TANDEM: Bi-Level Data Mixture Optimization with Twin Networks
  Jiaxing Wang, Deping Xiang, Jin Xu, Mingyang Yi, Guoqiang Gong, Zicheng Zhang, et al.
  NeurIPS 2025.

• Beyond Logits: Aligning Feature Dynamics for Effective Knowledge Distillation
  Guoqiang Gong, Jiaxing Wang, Jin Xu, Deping Xiang, Zicheng Zhang, Leqi Shen, et al.
  ACL 2025.

Generative Vision

• Resolving Endpoint Underfitting in Diffusion Bridges via Noise Alignment
  Yurong Gao*, Zicheng Zhang*, Congying Han, Tiande Guo, Xinmin Qiu.
  CVPR 2026.

• Focus on Neighbors and Know the Whole: Towards Consistent Dense Multiview Text-to-Image Generator for 3D Creation
  Bonan Li*, Zicheng Zhang*, Xingyi Yang, Xinchao Wang.
  CVPR 2025 Highlight.

• Transforming Radiance Field with Lipschitz Network for Photorealistic 3D Scene Stylization
  Zicheng Zhang, Yinglu Liu, Congying Han, Yingwei Pan, Tiande Guo, Ting Yao.
  CVPR 2023 Highlight.

• Towards Consistent Video Editing with Text-to-Image Diffusion Models
  Zicheng Zhang, Bonan Li, Xiaoyang Nie, Congying Han, Tiande Guo, Luoqi Liu.
  NeurIPS 2023.

• PetsGAN: Rethinking Priors for Single Image Generation
  Zicheng Zhang, Yinglu Liu, Congying Han, Hailin Shi, Tiande Guo, Bowen Zhou.
  AAAI 2022 Oral.

Machine Learning and Optimization

• Harnessing LLMs with Direction-Scope Control for Combinatorial Solver Refinement
  Yili Cao, Anqi Li*, Haoran Li, Yongtang Shi, Zicheng Zhang*.
  Preprint, 2026.

• On the Tension Between Optimality and Adversarial Robustness in Policy Optimization
  Haoran Li, Jiayu Lv, Congying Han, Zicheng Zhang, Anqi Li, Yan Liu, et al.
  ICLR 2026.

• Purity Law for Neural Routing Problem Solvers with Enhanced Generalizability
  Wenzhao Liu, Haoran Li, Congying Han, Zicheng Zhang, Anqi Li, Tiande Guo.
  NeurIPS 2025.

• Towards Optimal Adversarial Robust Q-learning with Bellman Infinity-error
  Haoran Li, Zicheng Zhang, Wang Luo, Congying Han, Yudong Hu, Tiande Guo, Shichen Liao.
  ICML 2024 Oral.

* Equal contribution or corresponding author.

For a complete publication list, please visit my Google Scholar profile.