Shikun Feng

Tsinghua University. BEIJING, P.R.China

prof_pic.jpg

Updated: I recently joined Zhongguancun Academy as an Assistant Professor and PhD supervisor. and I am currently recruiting PhD students. If you have already been admitted to Zhongguancun Academy and are interested in my research areas, please feel free to contact me via email at kunayumi@163.com or reach out to me directly on DingTalk.

My name is Shikun Feng, and I am a fourth-year PhD candidate in the Department of Computer Science at Tsinghua University, under the supervision of Professors Yanyan Lan and Wei-Ying Ma. My research interests include Multimodal Learning, Computer Vision, and AI for Science. Prior to pursuing my PhD, I worked as a senior computer vision algorithm engineer at SenseTime.

Recent Works

  1. arXiv
    LiveVVT: High-Fidelity Video Virtual Try-On in Real Time
    Yushe Cao, Shikun Feng, Ruxiang Duan, and 4 more authors
    arXiv preprint arXiv:2608.26714, 2026
    arXiv HTML
  2. arXiv
    LLaVA-OneVision-2: Towards Next-Generation Perceptual Intelligence
    Xiang An, Yin Xie, Feilong Tang, and 27 more authors
    arXiv preprint arXiv:2605.25979, 2026
    arXiv Code Website
  3. Neurocomputing
    Query-calibrated attention and visibility-aware anisotropic regression for small object detection
    Ruxiang Duan, Qiliang Du, Lianfang Tian, and 3 more authors
    Neurocomputing, 2026
    HTML
  4. arXiv
    OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining
    Zhongzheng Li, Tiancan Feng, Wenhao Li, and 5 more authors
    arXiv preprint arXiv:2607.20486, 2026
    arXiv HTML
  5. arXiv
    UniVVT: A Unified End-to-End Framework for High-Fidelity Video Virtual Try-on
    Yushe Cao, Shikun Feng, Fei Shen, and 5 more authors
    arXiv preprint arXiv:2608.05745, 2026
    arXiv HTML
  6. arXiv
    OneVision-Encoder: Codec-Aligned Sparsity as a Foundational Principle for Multimodal Intelligence
    Feilong Tang, Xiang An, Yunyao Yan, and 16 more authors
    arXiv preprint arXiv:2602.08683, 2026
    arXiv HTML Code Website
  7. ICML26
    MCCE: A Framework for Multi-LLM Collaborative Search in Discrete Spaces with Similarity-Filtered Preference Learning
    Nian Ran, Zhongzheng Li, Yue Wang, and 5 more authors
    In Forty-third International Conference on Machine Learning, 2026
    HTML PDF Code
  8. bioRxiv
    Ophiuchus-Ab: A Versatile Generative Foundation Model for Advanced Antibody-Based Immunotherapy
    Yiheng Zhu, Jian Ma, Mingze Yin, and 10 more authors
    bioRxiv, 2026
    HTML
  9. bioRxiv
    Functional in-context learning in genomic language models with nucleotide-level supervision and genome compression
    Qiuyi Li, Zhihao Zhan, Shikun Feng, and 11 more authors
    bioRxiv, 2026
    HTML

Selected Publications

  1. NeurIPS25
    FIGRDock: Fast Interaction-Guided Regression for Flexible Docking
    Shikun Feng, Bicheng Lin, Yuanhuan Mo, and 6 more authors
    2025
  2. NeurIPS25
    Straight-Line Diffusion Model for Efficient 3D Molecular Generation
    Yuyan Ni, Shikun Feng, Haohan Chi, and 5 more authors
    arXiv preprint arXiv:2503.02918, 2025
  3. ICLR25
    UniGEM: A Unified Approach to Generation and Property Prediction for Molecules
    Shikun Feng, Yuyan Ni, Yan Lu, and 3 more authors
    arXiv preprint arXiv:2410.10516, 2025
  4. ICML23
    Fractional denoising for 3d molecular pre-training
    Shikun Feng, Yuyan Ni, Yanyan Lan, and 2 more authors
    In International Conference on Machine Learning, 2023
    PDF
  5. arXiv
    Unimap: universal smiles-graph representation learning
    Shikun Feng, Lixin Yang, Weiying Ma, and 1 more author
    arXiv preprint arXiv:2310.14216, 2023
    HTML
  6. ICLR24
    Protein-ligand binding representation learning from fine-grained interactions
    Shikun Feng, Minghao Li, Yinjun Jia, and 2 more authors
    In The Twelfth International Conference on Learning Representations, 2024
    PDF
  7. ICML24
    UniCorn: A Unified Contrastive Learning Approach for Multi-view Molecular Representation Learning
    Shikun Feng, Yuyan Ni, Minghao Li, and 4 more authors
    In Forty-first International Conference on Machine Learning, 2024
    PDF
  8. Nat Mach Intell
    Pre-training with fractional denoising to enhance molecular property prediction
    Yuyan Ni, Shikun Feng, Xin Hong, and 5 more authors
    Nature Machine Intelligence, 2024
    HTML
  9. ICLR24
    Sliced Denoising: A Physics-Informed Molecular Pre-Training Method
    Yuyan Ni, Shikun Feng, Wei-Ying Ma, and 2 more authors
    In The Twelfth International Conference on Learning Representations, 2024
    PDF
  10. ICLR24
    Multimodal Molecular Pretraining via Modality Blending
    Qiying Yu, Yudi Zhang, Yuyan Ni, and 4 more authors
    In The Twelfth International Conference on Learning Representations, 2024
    PDF