đ About Me
I am currently pursuing a Ph.D. at the College of Information Science and Electronic Engineering (ISEE) at Zhejiang University under the supervision of Prof. Hui-liang Shen (Chair of the Department of Electronic Engineering), and Prof. Gang Pan, (Director of the State Key Laboratory of Brain-Machine Intelligence and Vice Dean of the College of Computer Science and Technology). Prior to this, I received postgraduate exemption and my Bachelorâs degree from the ISEE, Zhejiang University.
My research interests include 3D Reasoning and Neural Decoding, with VLMs. Recent research focuses on 4D-Radar-based VLA and fMRI-based semantic decoding with LLMs.
- Currently exploring the potential of large vision-language models in 3D spatial reasoning.
- Currently exploring brain-signal decoding for motor imagery with large vision-language models.
- Currently exploring the integration of VLA models with brain signals for robotic manipulation.
đ„ News
- Two paper submitted to AAAI (first author x1, corresponding author x1).
- Six paper submitted to ICLR (first author x3, corresponding author x2, co-first-author x1).
- One paper submitted to PR (corresponding author x1).
- One paper submitted to TVT (corresponding author x1).
- One paper submitted to TPAMI (co-first-author x1).
- Two paper accepted by RAL (corresponding author x1, co-first-author x1).
đ Educations
- 2023.09 - present, Ph.D. candidate in the College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou.
- 2019.09 - 2023.06, B.E. in the College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou.
đ» Projects
- 2026.03 - present, Collaboration with Postdoc Jiaxuan Chen on neural signal decoding and cross-modal alignment with VLMs).
- 2025.08 - 2026.03, ChinaEye Valley Co-operative Project on maternal and infant health risk assessment.
- 2025.12 - 2026.06, Collaboration with VIVO on 4D dynamic scene reconstruction under sparse-view, combining generative models.
- 2024.01 - 2026.06, National Key R&D Program of China, 4D radar object detection and tracking in autonomous driving.
- 2022.06 - 2023.12, Collaboration with Apple on image denoising based on transfer learning and generative models.
- 2021.06 - 2022.06, Student Research Project SRTP on face video heart rate signal measurement via rPPG.
đ Selected Publications
* Co-first author; â corresponding author.
âInfraWorld: Motion-Aware World Modeling for Roadside 4D Occupancy Forecastingâ, arXiv, 2026. [Paper]
Songkai Wang*, Xiaokai Bai*â , Lei Yang, and Hui-Liang Shen. (co-first author, corresponding author)
âFFBL-Coop: Association-Decoupled Cooperative 3D Multi-Object Trackingâ, arXiv, 2026. [Paper]
Haoxin Wu, Xiaokai Baiâ . (corresponding author)
âRCVLA: 4D Radar-Grounded Semantic Reasoning and Trajectory Arbitration for Autonomous Drivingâ, arXiv, 2026. [Paper]
Lianqing Zheng*, Xiaokai Bai*, Yixuan Luo, Runwei Guan, Minghao Liu, Zhiqiang Wei, Hui-Liang Shen, Xichan Zhu, and Zhixiong Maâ . (co-first author)
âSGDet3D++: Geometry-Grounded Semantics for 4D Radar and Camera 3D Object Detectionâ, arXiv, 2026. [Paper]
Xiaokai Bai, Zhenyu Fan, Lianqing Zheng, Songkai Wang, Si-Yuan Cao, and Hui-Liang Shen.
âRoadOcc Learns When to Persist, Transport, or Refresh Memory for Roadside Occupancy Predictionâ, arXiv, 2026. [Paper]
Xiaokai Bai, Lei Yang, Songkai Wang, Lianqing Zheng, Si-Yuan Cao, and Hui-Liang Shen.
âKCTF-Net: Kinematic Compensation Temporal Fusion for 4D Radar 3D Object Detectionâ, IEEE Robotics and Automation Letters (RAL), vol. 11, no. 10, pp. 12112-12119, 2026. [Paper]
Xingkai Jin, Guangxian Xu, Fei Ma, and Xiaokai Baiâ . (corresponding author)
âInfraOcc: An Infrastructure Occupancy Benchmark with Static-to-Dynamic Reasoningâ, arXiv, 2026. [Paper] 
Lei Yang*, Xiaokai Bai*, Boqi Liâ , Chunmian Lin, Li Wangâ , Ziying Song, Jiahuan Zhang, Enhui Ma, Haibao Yu, Jiaqi Ma, and Kaicheng Yuâ . (co-first author)
âRaGS: Unleashing 3D Gaussian Splatting from 4D Radar and Monocular Cues for 3D Object Detectionâ, IEEE/CVF Conference on Computer Vision & Pattern Recognition (CVPR), 2026. [Paper] 
Xiaokai Bai, Chenxu Zhou, Lianqing Zheng, Si-Yuan Cao, Jianan Liu, Xiaohan Zhang, Zhengzhuang Zhang, Hui-Liang Shen.
âBoosting Instance Awareness via Cross-View Correlation with 4D Radar and Camera for 3D Object Detectionâ, IEEE Transactions on Multimedia (TMM), 2026. [Paper] 
Xiaokai Bai, Lianqing Zheng, Si-Yuan Cao, Xiaohan Zhang, Zhe Wu, Beinan Yu, Fang Wang, Jie Bai, and Hui-Liang Shen.
âSGDet3D: Semantics and Geometry Fusion for 3D Object Detection Using 4D Radar and Cameraâ, IEEE Robotics and Automation Letters (RAL), vol. 10, no. 1, pp. 828-835, 2025. [Paper] 
Xiaokai Bai, Zhu Yu, Lianqing Zheng, Xiaohan Zhang, Zili Zhou, Xue Zhang, Fang Wang, Jie Bai, and Hui-Liang Shen.
â4DR360âŠ: State Reasoning for Joint 3D Detection and Occupancy Prediction in 4D Radar-Camera Full-Scene Perceptionâ, arXiv, 2026. [Paper]
Xiaokai Bai, Lianqing Zheng, Runwei Guan, Songkai Wang, Siyuan Cao, Hui-liang Shen.
âMindEdit-Bench: Benchmarking Object-Level Counterfactual Spatial Reasoning in VLMs from In-the-Wild Photosâ, arXiv, 2026. [Paper]
Leyuan Yu, Xiao Tang, Minghao Liu, Xinyuan Li, Xiaokai Bai, Sheng Zhou, Qunshu Lin, Weihao Xuan, and Naoto Yokoya.
âRC-GeoCP: Geometric Consensus for Radar-Camera Collaborative Perceptionâ, arXiv, 2026. [Paper]
Xiaokai Bai, Lianqing Zheng, Runwei Guan, Siyuan Cao, Songkai Wang, Huiliang Shen.
âLGDD: Local-Global Synergistic Dual-Branch 3D Object Detection Using 4D Radarâ, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2025. [Paper] 
Xiaokai Bai, Qin Yang, Zili Zhou, Fuyi Zhang, Zhe Wu, Si-Yuan Cao, Lianqing Zheng, Beinan Yu, Fang Wang, Jie Bai, and Hui-Liang Shen.
âSD4R: Sparse-to-Dense Learning for 3D Object Detection with 4D Radarâ, IEEE Intelligent Transportation Systems Conference (ITSC), 2025. [Paper] 
Xiaokai Bai, Jiahao Cheng, Songkai Wang, Yixuan Luo, Lianqing Zheng, Xiaohan Zhang, Si-Yuan Cao, and Hui-Liang Shen.
âOmniHD-Scenes: A Next-Generation Multimodal Dataset for Autonomous Drivingâ, IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI). [Paper] [Project] 
Lianqing Zheng, Long Yang, Qunshu Lin, Wenjin Ai, Minghao Liu, Shouyi Lu, Jianan Liu, Hongze Ren, Jingyue Mo, Xiaokai Bai, Jie Bai, Zhixiong Ma, and Xichan Zhu.
âDoracamom: Joint 3D Detection and Occupancy Prediction with Multi-view 4D Radars and Cameras for Omnidirectional Perceptionâ, IEEE Transactions on Circuits and Systems for Video Technology (TCSVT). [Paper] 
Lianqing Zheng, Jianan Liu, Runwei Guan, Long Yang, Shouyi Lu, Yuanzhe Li, Xiaokai Bai, Jie Bai, Zhixiong Ma, Hui-Liang Shen, and Xichan Zhu.
â Services
- Reviewer of CVPR, ECCV, AAAI, ACMMM, ICRA, IROS, ITSC, IV
- Reviewer of TCSVT, PR, IOTJ, TGRS, KBS, RAL, SensorsJ
đ Honors and Awards
- 2024-2025, Five-Good Graduate Student (top 1%)
- 2024-2025, Outstanding Graduate Student Cadre (top 5%)
- 2023-2024, Excellent Postgraduate student of Zhejiang University (top 5%)
- 2022-2023, Outstanding Graduate of Zhejiang Province (top 1%)
- 2022-2023, Postgraduate exemption based on being in the top 5%
- 2021-2023, Zhejiang University Student Research Training Program (SRTP), Outstanding
- 2021-2022, Zhejiang University Student Entrepreneurship Competition, Second Prize
- 2021-2022, Silver Award in the Zhejiang Internet+ Innovation and Entrepreneurship Competition
- 2021-2022, Zhejiang Provincial Government Scholarship (top 2%, ïż„6,000)
- 2021-2022, Honorable Mention in the Mathematical Contest in Modeling (MCM)
- 2021-2022, Zhejiang University Excellent Student (top 3%)
- 2020-2021, Zhejiang University First-Class Academic Scholarship (top 3%, ïż„6,000)
- 2020-2021, Zhejiang University Outstanding Senior Student
- 2019-2020, Zhejiang University First-Class Academic Scholarship (top 3%, ïż„6,000)
- 2019-2020, Zhejiang University Excellent League Cadre