Shanglin Li Homepage

Shanglin Li is a doctoral researcher with Klaus-Robert Müller at the Technical University of Berlin and a Research Associate with Mitsuo Kawato at ATR Japan. Shanglin is dedicated to developing modern machine learning techniques for neuroimaging, while maintaining interests in generic machine learning topics. Shanglin actively collaborates with world-leading labs in machine learning and neural engineering.

Shanglin’s research interests include geometric deep learning, transfer learning, multimodal neuroimaging, and brain foundation models.

Researchers in related research fields are welcome to collaborate.

Key Publications

• EEG-Based Multimodal Learning via Hyperbolic Mixture-of-Curvature Experts
  R Zhou^†, S Li^†, G Huang, X Zhou, Q Zhao, M Kawanabe, Y Ding, C Guan
  International Conference on Machine Learning (ICML), 2026.
  Equal contribution.

• HEEGNet: Hyperbolic Embeddings for EEG
  S Li, S Chu, O Koc, Y Ding, Q Zhao, M Kawanabe, Z Chen
  International Conference on Learning Representations (ICLR), 2026.

• SPDIM: Source-free Unsupervised Conditional and Label Shift Adaptation in EEG
  S Li, M Kawanabe, R Kobler
  International Conference on Learning Representations (ICLR), 2025.

Work Experience

• Doctoral Researcher (05/2026–Present)
  Berlin Institute for the Foundations of Learning and Data (BIFOLD), Germany
  Focus: AI for science
  Reference: Prof. Dr. Klaus-Robert Müller

• Cooperative Researcher (04/2023–Present)
  Advanced Telecommunications Research Institute International (ATR), Japan
  Focus: Machine learning for neuroimaging
  Reference: Prof. Dr. Mitsuo Kawato

• Associate Researcher (04/2025–03/2026)
  RIKEN Center for Advanced Intelligence Project (AIP), Japan
  Focus: Machine learning for neuroimaging
  Reference: Prof. Dr. Qibin Zhao