Zedong Wang王泽栋

I am a CSE PhD student at HKUST under Prof. Dan Xu. I interned at Apple AIML, working on multimodal understanding and MLLM post-training. I received my B.Eng. from HUST, where I was fortunate to be advised by Prof. Xinggang Wang.

Perhaps the only thing worth noting about me in AI is, in an era when peer review is being overwhelmed by LLMs, I've received some reviewer awards at a few leading venues, such as ICML, ICLR (2.6%), and ECCV (2.7%). And it just comes down to one thing: “Treat others as you would like to be treated.”

Email Scholar Twitter LinkedIn Ack. Echoes

News

  1. Three papers got accepted at NeurIPS 2026. Congrats to Yucheng and Siyong.

  2. Recognized as Silver Reviewer at ICML 2026.

  3. Started my research internship at Apple.

  4. One paper on image editing, CARE-Edit, was accepted at CVPR 2026. Congrats to Yucheng.

  5. Recognized as Outstanding Reviewer at BMVC 2025 (21.3%, 247/1,161).

  6. Received the Best Reviewer Award at UniReps 2025 (one of two awardees).

  7. One paper on multi-task learning, Rep-MTL, was accepted at ICCV 2025 as a Highlight (top 2.3%).

  8. Recognized as Notable Reviewer at ICLR 2025 (2.6%, 473/18,323).

  9. Recognized as Outstanding Reviewer at BMVC 2024 (19.3%, 166/860).

  10. Recognized as Outstanding Reviewer at ECCV 2024 (2.7%, 198/7,293).

Experience

  • Mar 2026 – Sep 2026

    Apple, AIML

    Research Intern · Host: Zizhen Wang

Research

My overarching goal is to build scalable multimodal foundation models that perceive, understand, and interact with the world as holistically as humans do. Previously, I have worked on (i) multi-task learning: Rep-MTL (ICCV'25 Highlight), VTP (NeurIPS'26); (ii) multimodal understanding and generation: CARE-Edit (CVPR'26); (iii) efficient model architectures: MogaNet (ICLR'24).

  1. SAFE-Hair: Scalp-Anchored Fields for Exportable Single-View Hair Reconstruction

    Yucheng Wang*, Zedong Wang*, Yuetong Wu, Yue Ma, Dan Xu

    Annual Conference on Neural Information Processing Systems (NeurIPS), 2026

    arXiv Project Hugging Face Video

  2. Virtual Task Prompting for Multi-Task Scene Understanding

    Zedong Wang*, Yucheng Wang*, Dan Xu

    Annual Conference on Neural Information Processing Systems (NeurIPS), 2026

    arXiv Project Hugging Face Video

  3. CARE-Edit: Condition-Aware Routing of Experts for Contextual Image Editing

    Yucheng Wang*, Zedong Wang*, Yuetong Wu, Yue Ma, Dan Xu

    IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026

    arXiv Project Hugging Face Video

  4. Rep-MTL: Unleashing the Power of Representation-level Task Saliency for Multi-Task Learning

    Zedong Wang, Siyuan Li, Dan Xu

    IEEE/CVF International Conference on Computer Vision (ICCV), 2025, Highlight (top 2.3%)

    arXiv Project Hugging Face Video

  5. MergeVQ: A Unified Framework for Visual Generation and Representation with Disentangled Token Merging and Quantization

    Siyuan Li*, Luyuan Zhang*, Zedong Wang, Juanxi Tian, Cheng Tan, Zicheng Liu, Chang Yu, Qingsong Xie, Haonan Lu, Haoqian Wang, Zhen Lei

    IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025

    arXiv Project Hugging Face

  6. Taming LLMs by Scaling Learning Rates with Gradient Grouping

    Siyuan Li*, Juanxi Tian*, Zedong Wang*, Xin Jin, Zicheng Liu, Wentao Zhang, Dan Xu

    Annual Meeting of the Association for Computational Linguistics (ACL), 2025

    arXiv Project Hugging Face

  7. MogaNet: Multi-order Gated Aggregation Network

    Siyuan Li*, Zedong Wang*, Zicheng Liu, Cheng Tan, Haitao Lin, Di Wu, Zhiyuan Chen, Jiangbin Zheng, Stan Z. Li

    International Conference on Learning Representations (ICLR), 2024

    arXiv Project

  8. Short-Long Convolutions Help Hardware-Efficient Linear Attention to Focus on Long Sequences

    Zicheng Liu, Siyuan Li, Li Wang, Zedong Wang, Yunfan Liu, Stan Z. Li

    International Conference on Machine Learning (ICML), 2024

    arXiv Project

  9. OpenMixup: Open Mixup Toolbox and Benchmark for Visual Representation Learning

    Siyuan Li*, Zedong Wang*, Zicheng Liu, Di Wu, Stan Z. Li

    arXiv preprint, 2022

    arXiv Project

* Equal contribution.

Awards

Reviewer awards

Recognitions

  • 2025

    Broadening Participation Award, ICCV 2025

  • 2025

    Highlight Presentation, ICCV 2025 (2.3%)

Service

Conference reviewer / PC member

  • International Conference on Learning Representations (ICLR) 2024 (Tiny Papers), 2025, 2026
  • Annual Conference on Neural Information Processing Systems (NeurIPS) 2024, 2025, 2026
  • International Conference on Machine Learning (ICML) 2024, 2025, 2026
  • IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2025, 2026
  • European Conference on Computer Vision (ECCV) 2024, 2026
  • AAAI Conference on Artificial Intelligence (AAAI) 2025
  • ACM International Conference on Multimedia (ACM MM) 2024
  • IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) 2026
  • British Machine Vision Conference (BMVC) 2024, 2025, 2026
  • International Conference on Pattern Recognition (ICPR) 2024
  • Northern Lights Deep Learning Conference (NLDL) 2026

Journal reviewer

  • IEEE Transactions on Image Processing (TIP)
  • Transactions on Machine Learning Research (TMLR)
  • IEEE Transactions on Knowledge and Data Engineering (TKDE)

Memberships

  • Student Member, BMVA 2025 – present
  • Member, CVF 2025 – present
  • Student Member, CCF 2024 – 2026
  • Student Member, CSIG 2023 – 2024

Teaching

  • COMP 2011 Programming with C++ Fall 2026
  • COMP 5212 Machine Learning Fall 2025

Acknowledgements

My research would not be possible without the support of my mentors, collaborators, and colleagues:

Beyond academia, I've been fortunate to meet some truly wonderful friends, especially during my six years in Shenzhen. I treasure the time we shared. It shaped my youth and, in more ways than I probably realize, still carries me today. If you're reading this, some of our memories live in Echoes. Just want to say take care and all the best! My door is always open, and the coffee's on me.

In particular, I want to thank Yiwei for the companionship over the years, from Wuhan to Hong Kong. I couldn't have stayed sane through it all without you.

And to xy: we may not quite qualify as friends since we knew each other only briefly and met just a handful of times, yet you showed me a different side of Beijing and a way of living that I'd always longed for. Those conversations were among my happiest moments. Best of luck with everything ahead!

Miscellanea

I enjoy brewing coffee, watching movies, going on runs, and aimless walks. Feel free to follow me on Strava.

Lately I've come to enjoy browsing record stores and craft beer bars, and chatting with the staff.

I'm also a fan of Chinese Football Boy.