Yian Zhao (赵祎安)

Master Student @ PKU

Vision-Language-Action Model, Visual Perception and Understanding

          [CV] 


Biography

Hi! I am Yian Zhao (Chinese name: 赵祎安), currently a Master student in Peking University (expected graduation in June 2026), supervised by Prof. Jie Chen. Before this, I received a B.S. degree in Dalian University of Technology.

My research interests mainly focus on Vision-Language-Action Model for embodied AI, and 2D/3D Visual Perception and Understanding , with a focus on optimizing model computational efficiency and designing practical human-computer interaction methods.

If you are interested in discussing or collaborating with me, please feel free to contact me via email.

Research Goal

The objective of my research is to explore efficient and universal algorithms for visual perception and understanding, and to construct practical interaction paradigms for multimodal open-world scenarios in order to enhance the performance and efficiency of Vision-Language Models (VLMs) and Vision-Language-Action Models (VLAs), thus promoting the implementation of Artificial General Intelligence in the physical world, such as environmental perception and interaction decision for embodied AI agents, and real-time perception and understanding for autonomous driving.

News

Education

Peking University (PKU), China

Master Student, supervised by Prof. Jie Chen.
GPA: 3.89 / 4.00, Google Scholar Citations: 1500+
Sep. 2023 - Jun. 2026 (expected)

Dalian University of Technology (DUT), China

Bachelor of Computer Science and Technology
GPA: 3.78 / 4.00 (Top 1%), Outstanding Graduates
Sep. 2019 - Jun. 2023

Publications | [Google Scholar]

Selected Papers
DETRs Beat YOLOs on Real-time Object Detection 🔥🔥🔥 (Citations: 1300+)
Yian Zhao, Wenyu Lv, Shangliang Xu, Jinman Wei, Guanzhong Wang, Qingqing Dang, Yi Liu, Jie Chen

CVPR 2024

issues / [Arxiv] / [Project] / [Video]
iSegMan: Interactive Segment-and-Manipulate 3D Gaussians
Yian Zhao, Wanshi Xu, Ruochong Zheng, Pengchong Qiao, Chang Liu, Jie Chen

CVPR 2025

[Arxiv] / [Project] / [Video]
GraCo: Granularity-Controllable Interactive Segmentation
Yian Zhao, Kehan Li, Zesen Cheng, Pengchong Qiao, Xiawu Zheng, Rongrong Ji, Chang Liu, Li Yuan, Jie Chen

CVPR 2024 Highlight (Top 2.5%)

[Arxiv] / [Project] / [Video]
RT-DETRv2: Improved Baseline with Bag-of-Freebies for Real-Time Detection Transformer
Wenyu Lv, Yian Zhao, Qinyao Chang, Kui Huang, Guanzhong Wang, Yi Liu

Technical Report

[Arxiv] / [Project] / [Video]
Multi-granularity Interaction Simulation for Unsupervised Interactive Segmentation
Kehan Li, Yian Zhao, Zhennan Wang, Zesen Cheng, Peng Jin, Xiangyang Ji, Li Yuan, Chang Liu, Jie Chen

ICCV 2023

[Arxiv] / [Project]
ACSeg: Adaptive Conceptualization for Unsupervised Semantic Segmentation
Kehan Li, Zhennan Wang, Zesen Cheng, Runyi Yu, Yian Zhao, Guoli Song, Chang Liu, Li Yuan, Jie Chen

CVPR 2023 Highlight (Top 2.5%)

[Arxiv]
Blockchain-Enhanced Federated Learning Market With Social Internet of Things
Pengfei Wang, Yian Zhao, Mohammad S. Obaidat, Zongzheng Wei, Heng Qi, Chi Lin, Yunming Xiao, Qiang Zhang

IEEE Journal on Selected Areas in Communications (JSAC, JCR Q1, IF=13.8)

[Paper]
Federated Unlearning With Momentum Degradation
Yian Zhao, Pengfei Wang, Heng Qi, Jianguo Huang, Zongzheng Wei, Qiang Zhang

IEEE Internet of Things Journal (JCR Q1, IF=8.2)

[Paper]

Internship & Research Experience

IDEA 研究院

Mar. 2025 - Present, Research Intern

Supervised by Prof. Lei Zhang

Tencent AI Lab

Apr. 2024 - Mar. 2025, Research Intern

Supervised by Dr. Yang Wu and Dr. Jun Zhang

Baidu

Nov. 2022 - Jun. 2023, Research Intern

Working with Res. Wenyu Lv

Competition Awards (Selected)

16th "Challenge Cup" Liaoning Province College Students' Extracurricular Academic Science and Technology Works Competition

第十六届"挑战杯"辽宁省大学生课外学术科技作品竞赛

Grand Prize 🏆

With Haojun Tang et al. Supervised by Prof. Pengfei Wang.
Huawei Developer Competition Autonomous Vehicle Challenge National Finals

华为开发者大赛无人车挑战赛全国总决赛

Winner 🏆

With Ziyu Yue, Ruixi You.

Honors

Clubs

  • 2022 - Present, PPDE (PaddlePaddle Developers Experts) (飞桨开发者技术专家, PaddlePaddle is Baidu's open-source deep learning platform)
  • 2022 - 2023, Founding leader of Dalian University of Technology PaddleClub (飞桨领航团)
  • 2023 - Present, Openl 启智社区, 荣获2023年积极贡献者奖
  • Reports

    [5/2023], 河南大学人工智能学院百度飞桨校园AI DAY主题讲座
    [11/2022], 《挺进热门专业的学生,怎样用飞桨“卷”到保研机会?》

    Latest updated in Mar. 2025

    This cool template is stolen from Tianxing Chen