Zihan Weng

PhD student in the Integrated Program in Neuroscience at McGill University, advised by Pouya Bashivan. I study how biological memory systems can inspire efficient and scalable deep learning architectures.

ZW
McGill UniversityMilaBashivanLab
Email CV / Resume
Dec 2025

Instructor for Cognitive Benchmarking in Large Models workshop at MAIN 2025, Montreal.

Sep 2025

NeurIPS 2025 Spotlight: Caption This, Reason That: VLMs Caught in the Middle.

Sep 2025

Started PhD at McGill University in the Integrated Program in Neuroscience.

Jun 2025

Received Excellent Master's Thesis Award (Top 2%) from UESTC.

🧠

Brain-Inspired Architectures

Drawing on the brain's hierarchical memory systems to design more efficient sequence models, with current focus on linear recurrent architectures.

🔍

Cognitive Evaluation of AI

Applying paradigms from cognitive psychology to systematically evaluate and expose failure modes in large multimodal and reasoning models.

🤖

Neural Signal Decoding

Developing deep learning methods for real-time decoding of fine motor intentions from surface electromyography signals for prosthetic control.

Education

  • McGill University, Montreal, Canada

    		2025 – Present
    Doctor of Philosophy in the Integrated Program in Neuroscience
    Advisor: Prof. Pouya Bashivan

  • McGill University, Montreal, Canada

    		2023 – 2025
    Master of Science in the Integrated Program in Neuroscience, Joint dual-degree program with UESTC
    Research Focus: Evaluation and Fine-tuning of Large Multimodal Models
    Advisor: Prof. Pouya Bashivan

  • University of Electronic Science and Technology of China (UESTC), Chengdu, China

    		2022 – 2025
    Master of Science in Biomedical Engineering, Joint dual-degree program with McGill University
    Research Focus: EMG-based Hand Fine Motor Decoding using Deep Learning
    Advisor: Prof. Peng Xu

  • Northeast Electric Power University (NEEPU), Jilin, China

    		2018 – 2022

    Bachelor of Engineering in Computer Science

Selected Publications

  • Zihan Weng, Lucas Gomez, Taylor Whittington Webb, and Pouya Bashivan. “Caption This, Reason That: VLMs Caught in the Middle”. In: The Thirty-ninth Annual Conference on Neural Information Processing Systems. Spotlight, Poster. 2025
  • Zihan Weng, Yang Xiao, Peiyang Li, Chanlin Yi, Pouya Bashivan, Hailin Ma, Guang Yao, Yuan Lin, Fali Li, Dezhong Yao, Jingming Hou, Yangsong Zhang, and Peng Xu. “Real-time fine finger motion decoding for transradial amputees with surface electromyography”. In: Neural Networks (2025), p. 107605. issn: 0893-6080. doi: https://doi.org/10.1016/j.neunet.2025.107605
  • Zihan Weng, Xiabing Zhang, Yufeng Mou, Chanlin Yi, Fali Li, Pouya Bashivan, and Peng Xu. “A Hybrid CNN-Transformer Approach for Continuous Fine Finger Motion Decoding from sEMG Signals”. In: 2024 IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications (CIVEMSA) (2024). Oral, pp. 1–6
  • Xiabing Zhang, Chen Liu, Zihan Weng, Bin Lu, Fali Li, and Peng Xu. “The corticomuscular coupling underlying movement and its application for rehabilitation: a review”. In: Brain-Apparatus Communication: A Journal of Bacomics 2.1 (2023), p. 2183096. doi: 10.1080/27706710.2023.2183096. eprint: https://doi.org/10.1080/27706710.2023.2183096
  • Xiabing Zhang, Bin Lu, Zihan Weng, Yifeng Wang, Jingming Hou, Jing Qiu, Dezhong Yao, Fali Li, and Peng Xu. “Within and between electrophysiology networks for EEG and EMG under different thumb forces”. In: Biomedical Signal Processing and Control 86 (2023), p. 105249. issn: 1746-8094. doi: https://doi.org/10.1016/j.bspc.2023.105249
  • Chanlin Yi, Jiamin Zhang, Zihan Weng, Wanjun Chen, Dezhong Yao, Fali Li, Zehong Cao, Peiyang Li, and Peng Xu. “Nonparametric Dynamic Granger Causality based on Multi-Space Spectrum Fusion for Time-varying Directed Brain Network Construction”. In: IEEE Journal of Biomedical and Health Informatics (2024), pp. 1–10. doi: 10.1109/JBHI.2024.3477944

Teaching & Workshops

  • Cognitive Benchmarking in Large Models, MAIN 2025 (Montreal AI & Neuroscience Conference)

    		Dec 2025
  • Co-designed and delivered a hands-on workshop covering LLMs, Vision-Language Models, and reasoning models.

Awards and Honors

  • Full Doctoral Fellowship, China Scholarship Council ($105,600)

    		2025 – 2029

  • IPN Graduate Excellence Fellowship ($15,000)

    		2025 – 2028

  • Excellent Master’s Thesis Award (Top 2%), UESTC

    		2025

  • National Scholarship (Top 0.7% of Chinese undergraduate students), Ministry of Education, China

    		2020

  • Presidential Scholarship (10 recipients out of 15,000 students), NEEPU

    		2021

Technical Skills

  • Programming: Python, C/C++, JavaScript
  • ML Stack: PyTorch, HuggingFace, scientific Python ecosystem
  • Systems: Linux, Git, SLURM/HPC, LaTeX
  • Languages: Chinese (native), English