Biography

Education

• Columbus, OH Ohio State University (OSU)
  2022 – 2026 Doctor of Philosophy in Computer Science and Engineering
• Ann Arbor, MI University of Michigan (UMich)
  2019 – 2021 Master of Science in Electrical Engineering and Computer Science
• Wuhan, China Huazhong University of Science and Technology (HUST)
  2015 – 2019 Bachelor of Science in Electrical Engineering and Automation

Experience

• Senior Researcher @ Microsoft, Redmond, WA Feb. 2026 – Present
• Machine Learning Engineer Intern @ Adobe, San Jose, CA May 2024 – Aug. 2024
• Research Intern @ Microsoft, Redmond, WA May 2023 – Aug. 2023
• Research Assistant @ Ohio State University, Columbus, OH Aug. 2022 – Jan. 2026
• Research Assistant @ University of Michigan, Ann Arbor, MI Jan. 2020 – Dec. 2020

Selected Publications

^* indicates equal contribution

1. Improving Visual Discriminability of CLIP for Training-Free Open-Vocabulary Semantic Segmentation
   Jinxin Zhou, Jiachen Jiang, and Zhihui Zhu
   TMLR, 2026
   
2. Analyzing Fine-Grained Alignment and Enhancing Vision Understanding in Multimodal Language Models
   Jiachen Jiang, Jinxin Zhou, Bo Peng, Xia Ning, and Zhihui Zhu
   NeurIPS, 2025 AC rate = 24.5%
   
3. On Layer-wise Representation Similarity: Application for Multi-Exit Models with a Single Classifier
   Jiachen Jiang, Jinxin Zhou, Zhihui Zhu
   ICLR, 2025 AC rate = 32%
   
4. A Validation Approach to Over-parameterized Matrix and Image Recovery
   Lijun Ding, Zhen Qin, Liwei Jiang, Jinxin Zhou, and Zhihui Zhu
   CPAL, 2025
   
5. Are All Layers Created Equal: A Neural Collapse Perspective
   Jinxin Zhou, Jiachen Jiang, and Zhihui Zhu
   CPAL, 2025
   
6. DREAM: Diffusion Rectification and Estimation-Adaptive Models
   Jinxin Zhou^*, Tianyu Ding^*, Tianyi Chen, Jiachen Jiang, Ilya Zharkov, Zhihui Zhu, and Luming Liang
   CVPR, 2024 AC rate = 23.6%
   pdf code arXiv project page
7. Generalized Neural Collapse for A Large Number of Classes
   Jiachen Jiang^*, Jinxin Zhou^*, Peng Wang, Qing Qu, Dustin Mixon, Chong You, and Zhihui Zhu
   ICML, 2024 AC rate = 27.5%
   
8. Principled and Efficient Transfer Learning of Deep Models via Neural Collapse
   Xiao Li, Sheng Liu, Jinxin Zhou, Xinyu Lu, Fernandez-Granda Carlos, Zhihui Zhu, and Qing Qu
   TMLR, 2023
   
9. Are All Losses Created Equal: A Neural Collapse Perspective
   Jinxin Zhou, Chong You, Xiao Li, Kangning Liu, Sheng Liu, Qing Qu, and Zhihui Zhu
   NeurIPS, 2022 AC rate = 25.6%
   
10. On the Optimization Landscape of Neural Collapse under MSE Loss: Global Optimality with Unconstrained Features
    Jinxin Zhou^*, Xiao Li^*, Tianyu Ding, Chong You, Qing Qu, and Zhihui Zhu
    ICML, 2022 AC rate = 21.9%
    pdf code arXiv
11. A Geometric Analysis of Neural Collapse with Unconstrained Features Spotlight, top 3%
    Zhihui Zhu^*, Tianyu Ding^*, Jinxin Zhou, Xiao Li, Chong You, Jeremias Sulam, and Qing Qu
    NeurIPS, 2021 AC rate = 25.7%
    
12. Deep Learning Based Similarity-consistency Abnormality Detection (SCAD) Model for Classification of MRI Patterns of Multiple Myeloma Infiltration
    Jinxin Zhou, Chuan Zhou, Heang-Ping Chan, Lubomir M Hadjiiski, and Qian Dong
    SPIE Medical Imaging, 2021
    

Selected Preprints

1. On the Convergence of Gradient Descent on Learning Transformers with Residual Connections
   Zhen Qin^*, Jinxin Zhou^*, and Zhihui Zhu
   arXiv, 2025
   
2. From Compression to Expansion: A Layerwise Analysis of In-Context Learning
   Jiachen Jiang, Yuxin Dong, Jinxin Zhou, and Zhihui Zhu
   arXiv, 2025
   
3. D-STAR: Diffusion Model with Spatial-Temporal Adaptive Rectification
   Yong Ma, Luming Liang, Tianyu Ding, Jinxin Zhou
   arXiv, 2025
   
4. Cat-AIR: Content and Task-Aware All-in-One Image Restoration
   Jiachen Jiang, Tianyu Ding, Ke Zhang, Jinxin Zhou, Tianyi Chen, Ilya Zharkov, Zhihui Zhu, and Luming Liang
   arXiv:2503.17915, 2025
   
5. The Efficiency Spectrum of Large Language Models: An Algorithmic Survey
   Tianyu Ding, Tianyi Chen, Haidong Zhu, Jiachen Jiang, Yiqi Zhong, Jinxin Zhou, Guangzhi Wang, Zhihui Zhu, Ilya Zharkov, and Luming Liang
   arXiv:2312.00678, 2023
   
6. AdaContour: Adaptive Contour Descriptor with Hierarchical Representation
   Tianyu Ding, Jinxin Zhou, Tianyi Chen, Zhihui Zhu, Ilya Zharkov, and Luming Liang
   arXiv:2404.08292, 2024
   

Patents

• Jinxin Zhou, Hui Qu, Saeid Mottian, Krishna Kumar Singh, and Ratheesh Kalarot, Boximotion: Still Images into Dynamic Videos with Bounding Box Control, Adobe, 2024. (Submitted)

Professional Services

• Conference Reviewer:
  ICML (2022–2025), NeurIPS (2023–2025), ICLR (2022–2025), CVPR (2024–2025), ICASSP (2023)
• Journal Reviewer:
  IEEE TPAMI, JMLR, TMLR

Visitors