Shijing Wang
About Me
I am a Ph.D. candidate in Computer Science and Technology at Beijing Jiaotong University (BJTU), supervised by Prof. Yaping Huang. Since November 2025, I have been a CSC-funded visiting Ph.D. student at the University of Birmingham, supervised by Prof. Hyung Jin Chang, working on multimodal foundation models and human gaze reasoning.
My research focuses on multimodal models for human gaze understanding. I study gaze estimation and gaze following, uncertainty-aware and domain-generalizable learning, and vision-language/foundation models for attention and intention reasoning.
I am always open to academic collaboration and discussions. Please find my Chinese academic CV, Google Scholar, and DBLP profiles here.
News
- [2026-05] 🔥 A new preprint, Enhancing Gaze Reasoning in Vision Foundation Models for Gaze Following, is available on arXiv.
- [2026-04] 🔥 See Through the Noise: Improving Domain Generalization in Gaze Estimation is accepted to CVPR 2026.
- [2026-03] 🔥 CineSRD: Leveraging Visual, Acoustic, and Linguistic Cues for Open-World Visual Media Speaker Diarization is accepted to CVPR 2026.
- [2026-02] 🔥 Hermes the Polyglot is accepted to The Web Conference (WWW) 2026, and From Utterance to Vividity is accepted to ICLR 2026.
- [2025-12] Our work VL4Gaze: Unleashing Vision-Language Models for Gaze Following is available on arXiv.
- [2025-11] I joined the University of Birmingham as a CSC-funded visiting Ph.D. student.
- [2025-08] Fine-grained Video Dubbing Duration Alignment with Segment Supervised Preference Optimization was published at ACL 2025.
- [2024-06] 🏆 Our team won 1st place in the Ego4D Looking-at-Me Challenge at CVPR 2024.
- [2024-02] Suppressing Uncertainty in Gaze Estimation was published at AAAI 2024.
Education
- [Sep. 2021-present, integrated M.Eng.-Ph.D. program] Beijing Jiaotong University, Computer Science and Technology. Supervisors: Prof. Liping Jing (M.Eng. stage) and Prof. Yaping Huang (Ph.D. stage).
- [Sep. 2017-Jun. 2021, B.Eng.] Beijing Jiaotong University, Computer Science and Technology. Ranked in the top 20% and received direct admission to the master’s program.
Research Interests
- Gaze perception and visual attention: gaze estimation, gaze following, attention target and intention understanding.
- Vision-language and foundation models: multimodal instruction tuning, visual reasoning, and parameter-efficient adaptation.
- Reliable machine learning: uncertainty estimation, label-noise robustness, domain generalization, and evidential learning.
Research Experience
- University of Birmingham, School of Computer Science — CSC Visiting Ph.D. Student, Nov. 2025-present Research on gaze reasoning in vision foundation models and gaze following with vision-language models, including parameter-efficient adaptation, gaze-aware constraints, multimodal instruction data construction, and evaluation.
- Beijing Jiaotong University, Key Laboratory of Big Data & Artificial Intelligence in Transportation — Ph.D. Researcher, Sep. 2023-present Research on reliable gaze estimation, cross-dataset generalization, uncertainty modeling, and multimodal human attention understanding.
- Lenovo Research, Beijing — Machine Learning Research Intern, Dec. 2023-Jul. 2024 Worked on cross-dataset gaze estimation and first-person Looking-at-Me recognition; contributed to the winning solution of the CVPR 2024 Ego4D challenge.
Publications
* denotes equal contribution. Shijing Wang is highlighted in bold.
Shijing Wang, Yaping Huang, Chaoqun Cui, David Wong, Yihua Cheng, Alexandros Neophytou, Hyung Jin Chang
arXiv: 2605.22607
Shijing Wang*, Chaoqun Cui*, Yaping Huang, Hyung Jin Chang, Yihua Cheng
arXiv: 2512.20735
[AAAI 2024] Suppressing Uncertainty in Gaze Estimation
Shijing Wang, Yaping Huang
Proceedings of the AAAI Conference on Artificial Intelligence, 38(6), 5581-5589 · CCF-A
Shijing Wang, Yaping Huang, Jun Xie, Yi Tian, Feng Chen, Zhepeng Wang
arXiv: 2409.04766
Yanming Peng, Shijing Wang, Yaping Huang, Yi Tian
Accepted · CCF-A
[CVPR 2026] CineSRD: Leveraging Visual, Acoustic, and Linguistic Cues for Open-World Visual Media Speaker Diarization
Liangbin Huang*, Xiaohua Liao*, Chaoqun Cui*, Shijing Wang, Zhaolong Huang, Yanlong Du, Wenji Mao
Accepted · CCF-A
[ACL 2025] Fine-grained Video Dubbing Duration Alignment with Segment Supervised Preference Optimization
Chaoqun Cui, Liangbin Huang, Shijing Wang, Zhe Tong, Zhaolong Huang, Xiao Zeng, Xiaofeng Liu
Main conference, pp. 4524-4546 · CCF-A
Chaoqun Cui, Shijing Wang, Liangbin Huang, Qingqing Gu, Zhaolong Huang, Xiao Zeng, Wenji Mao
Accepted · CCF-A
Chaoqun Cui, Shijing Wang, Liangbin Huang, Qingqing Gu, Zhaolong Huang, Xiao Zeng, Wenji Mao
Accepted, pp. 7068-7079 · CCF-A
Chaoqun Cui, Jing Huang, Shijing Wang, Liming Zheng, Qingchao Kong, Zhixiong Zeng
arXiv: 2602.00575
Xin Liu, Shijing Wang, Kairui Zhou, Yilin Lyu, Mingyang Song, Liping Jing, Tieyong Zeng, Jian Yu
Research Track, pp. 454-470 · CCF-B
[CVPR 2024 Challenge] PCIE_LAM Solution for Ego4D Looking At Me Challenge
Kanokphan Lertniphonphan, Jun Xie, Yaqing Meng, Shijing Wang, Feng Chen, Zhepeng Wang
1st place in the Ego4D Looking-at-Me Track
Projects and Awards
- Research on Gaze Estimation Generalization, Fundamental Research Funds for the Central Universities, Principal Investigator, Apr. 2024-Mar. 2026.
- Chinese Utility Model Patent, A Stacked Connector Welding Plate (一种叠式连接器焊接板), CN211184400U, granted in 2020. Inventors: Shijing Wang, Xin Gao, Hua Lu, and Yongping Teng. (patent)
- China Scholarship Council (CSC) Scholarship, funded visiting Ph.D. study at the University of Birmingham, Nov. 2025-present.
- 1st Place, Ego4D-Ego-Exo4D Challenge, Looking-at-Me Track, CVPR 2024 (code, paper).
- First-Class Academic Scholarship, Beijing Jiaotong University, 2020-2024 (five times).
- Outstanding Undergraduate Thesis, Beijing Jiaotong University, 2021.
- National Undergraduate Innovation Training Program, 2020.