Awards & Honours

  • 2026 Rising Star Award, Conference on Parsimony and Learning (CPAL 2026)
  • 2022 Best Paper Award, Learning on Graphs Conference (LoG 2022) — Untrained GNNs Tickets
  • 2017 Best Paper Nomination Award, International Conference on Computer Vision Systems (ICVS 2017)

Invited Talks

  • 2026 Where to Spend Parameters: From Layerwise Efficiency to Federated Architecture Search — Tutorial at CPAL 2026 (slides)
  • 2025 Layerwise Insights: A Secret Sauce for LLM Efficiency, Tsinghua University
  • 2024 The Power of Model Sparsity, Multimedia Analytics (MA) Laboratory, City University of Hong Kong
  • 2023 LLM Pruning, Visual Informatics Group, University of Texas at Austin
  • 2023 Meta Universe and Digital Human, AI Time PhD Debate, Tsinghua University
  • 2022 Model / Supervision Efficiency, Xu Lab, Carnegie Mellon University
  • 2020 Going Beyond Training ML Models with Labels, EDGE AI, Eindhoven University of Technology

Tutorials

  • 2026 Where to Spend Parameters: From Layerwise Efficiency to Federated Architecture Search — Tutorial at CPAL 2026 (slides)

Associate Editor

  • Visual Intelligence (Link)

Senior Program Committee

  • 2026 Area Chair of ICPR (International Conference on Pattern Recognition)
  • 2026 Area Chair of NeurIPS (Conference on Neural Information Processing Systems)
  • 2025 Area Chair of CPAL (Conference on Parsimony and Learning)

Program Committee / Reviewer

  • 2026 ICML (Silver Reviewer), NeurIPS
  • 2025 CPAL, ICLR, CVPR, BMVC, NeurIPS
  • 2024 NeurIPS, UAI, ICML, CPAL
  • 2023 NeurIPS, UAI, ICLR, SNN Workshop

Journal Invited Reviewer

  • ACM Computing Surveys
  • IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
  • Neurocomputing
  • IEEE Signal Processing Letters
  • Clinical Epidemiology

Organizing Committee

  • Publicity Chair, International Conference on Artificial Intelligence and Agents (Link)
  • NeurIPS 2024 Challenge: Edge-Device Large Language Model Competition (Link)
  • IEEE CAI 2025 Workshop: Stable Training Paradigms for LLMs — Reducing Instability, Increasing Capacity (Link)
  • IEEE CAI 2025 Workshop: Secure, Private, and Fair Federated Optimization and Learning (Link)