About Me

Hi! I’m Yang Liu (柳洋, E-mail: liuyang232@mails.ucas.ac.cn). Now I am a PhD. student of School of Computer Science and Technology, University of Chinese Academy of Sciences (SCST, UCAS), supervised by Prof. Qingming Huang (黄庆明) (IEEE Fellow) and Qianqian Xu (许倩倩) (Professor at ICT, CAS). I have received the B.E. degree in Software Engineering from HIT in 2023. My research interest includes self-supervised learning, video understanding, and computer vision.

🔥 News

  • 2026.02:  🎉🎉 One paper has been accepted by CVPR 2026.
  • 2025.09:  🎉🎉 One paper has been accepted by NeurIPS 2025.
  • 2025.06:  🎉🎉 Winning two first-prizes of the CVPR 2025 Challenges.
  • 2025.02:  🎉🎉 One paper has been accepted by CVPR 2025.
  • 2024.10:  🎉🎉 Our ACM MultiMedia 2024 paper has been presented as Honourable Mention Award!
  • 2024.07:  🎉🎉 One paper has been accepted by ACM MultiMedia 2024 and further selected as an oral.

📝 Publications

CVPR 2026
sym

From Static to Dynamic: Exploring Self-supervised Image-to-Video Representation Transfer Learning

Yang Liu, Qianqian Xu*, Peisong Wen, Siran Dai, Xilin Zhao, Qingming Huang*. Conference on Computer Vision and Pattern Recognition (CVPR, CCF-A) 2026. | [Code] | [Slides] | [Project]

  • This paper proposes a self-supervised image-to-video transfer learning method for effective trade-off between intra-video temporal consistency and inter-video semantic separability.
CVPR 2025
sym

When the Future Becomes the Past: Taming Temporal Correspondence for Self-supervised Video Representation Learning

Yang Liu, Qianqian Xu*, Peisong Wen, Siran Dai, Qingming Huang*. Conference on Computer Vision and Pattern Recognition (CVPR, CCF-A) 2025. | [Code] | [Slides] | [Project]

  • This paper proposes a self-supervised framework for dense-level video representation learning with temporal correspondence.
ACMMM 2024 (Oral)
sym

Not All Pairs are Equal: Hierarchical Learning for Average-Precision-Oriented Video Retrieval

Yang Liu, Qianqian Xu*, Peisong Wen, Siran Dai, Qingming Huang*. ACM International Conference on Multimedia (ACM MultiMedia, CCF-A) 2024. (Honourable Mention Award) (Oral, 3.97%) | [Code] | [Slides] | [Project]

  • This paper proposes a self-supervised video retrieval framework with hierarchical AP optimization from the view of aligning the training objective and the evaluation metric.
NeurIPS 2025
sym

Exploring Structural Degradation in Dense Representations for Self-supervised Learning

Siran Dai, Qianqian Xu*, Peisong Wen, Yang Liu, Qingming Huang*. Annual Conference on Neural Information Processing Systems (NeurIPS, CCF-A) 2025. | [Code] | [Slides] | [Project]

  • This paper provides an observation of dense degradation phenomenon in self-supervised learning and introduces a Dense representation Quality Estimator to mitigate dense degradation.
CJC 2026
sym

Collaborative Salient Object Detection via Complex Background Semantic Modeling

Zitai Wang, Qianqian Xu*, Yuchen Cao, Yang Liu, Qingming Huang*. Chinese Journal of Computers (计算机学报, CCF-A) 2026. | [Code] | [Slides] | [Project]

  • This paper introduces a co-salient object detection method that improves detection accuracy by explicitly addressing the semantic ambiguity and heterogeneity of background regions.
CVPR 2026
sym

Safe Codebook: Token-Level Moderation for Safer Visual Autoregressive Generation

Jiaxuan Zhang, Qianqian Xu*, Peisong Wen, Siran Dai, Yang Liu, Qingming Huang*. Conference on Computer Vision and Pattern Recognition (CVPR Findings, CCF-A) 2026. | [Code] | [Slides] | [Project]

  • This paper proposes a training-free controllable framework for Visual Autoregressive models to effectively suppress harmful content while maintaining high fidelity to the original image.

💬 Invited Talks

  • 2025.11: Talk on University of Chinese Academy of Sciences 2025 Graduate Academic Forum. [Website]
  • 2025.10: Talk on third-place solution for ICCV 2025 Physics-IQ × Perception Test Challenge. [Slides]
  • 2025.10: Talk on 2025 China Journal of Image and Graphics Graduate Student Academic Salon. [Website]
  • 2025.06: Talk on first-place solution for CVPR 2025 AVA Challenge of 3D Human Motion Generation Track. [Slides]
  • 2024.12: Talk on Young Scientists Conference of CSIG 2024. [Poster]
  • 2024.10: Oral presentation of ACM MultiMedia 2024. [Slides]

🏅 Honors and Awards

Research Awards

  • 2025.10: Third place in ICCV 2025 Physics-IQ × Perception Test Challenge. [Website].
  • 2025.10: Outstanding report on 2025 China Journal of Image and Graphics Graduate Student Academic Salon. [Website]
  • 2025.06: First place in CVPR 2025 AVA Challenge of 3D Human Motion Generation Track. [Website].
  • 2025.06: First place in CVPR 2025 Computer Vision for Drug Discovery Challenge. [Website].
  • 2024.12: Excellent Poster on Young Scientists Conference of CSIG 2024. [Website].
  • 2024.10: ACM MultiMedia 2024 Honourable Mention Award. (ACM MultiMedia 2024 最佳论文荣誉提名奖) [Website].

Honorary Prizes

  • 2026.01: Excellent Students of Beijing. (北京市三好学生)
  • 2025.12: National Scholarship for Doctoral Students. (博士研究生国家奖学金)
  • 2025.05: President of Chinese Academy of Sciences Excellence Award. (中国科学院院长优秀奖)
  • 2025.05: Pacemaker to Merit Student of UCAS. (三好学生标兵)
  • 2023.07: Excellent Graduates of Shandong Province. (山东省优秀毕业生)
  • 2023.05: Excellent Students of Shandong Province. (山东省优秀学生)
  • 2023.04: The 21st Top Ten Undergraduates of HIT. (哈工大威海十佳大学生)
  • 2020.12: National Scholarship for Undergraduates. (本科生国家奖学金)

📖 Academic Services

Journal

  • IEEE Transactions on Circuits and Systems for Video Technology (IEEE T-CSVT): Reviewer
  • Multimedia Systems: Reviewer
  • Artificial Intelligence In Medicine: Reviewer

Conference

  • ACM MultiMedia: Reviewer (2024)
  • CVPR: Reviewer (2026)
  • ECCV: Reviewer (2026)

🎓 Educations

sym

2023.09 - Now, Ph.D. Student.
School of Computer Science and Technology.
University of Chinese Academy of Sciences (UCAS), Beijing.

sym

2019.09 - 2023.06, Undergraduate.
School of Computer Science and Technology.
Harbin Institute of Technology (HIT), Weihai.