Jiahang Li
Hi, I am Jiahang. I am an incoming Ph.D. student in Artificial Intelligence at Korea University, advised by Prof. Seong-Whan Lee , and a member of the PR&ML Lab . Previously, I was a research intern at the University of Electronic Science and Technology of China (UESTC), advised by Prof. Chaoning Zhang .
My research interests include multimodal foundation models, spanning representation learning and generative modeling for content creation. I am particularly interested in developing domain-specialist agents that enhance productivity, well-being, and creativity.
I am also the co-founder of JoyVoice Studio (noncommercial creative studio), which produces voice dubbing, original audio dramas, and original songs. I aim to draw on my experience in voice dubbing and sound design to strengthen multimodal understanding and advance human–AI collaboration in content creation.
I am always open to discussing new research ideas and potential collaborations. Please feel free to reach out.
Stay Hungry. Stay Foolish.
News
-
2026
-
🤗I will be working as a research assistant at the University of Electronic Science and Technology of China.
-
-
2025
-
😻 Honored to be recognized as an Outstanding Reviewer for the BMVC 2025 Conference.💖 Honor
-
📘 The first open-source comprehensive guide for deploying federated learning on NVIDIA Jetson AGX Xavier is now available on GitHub. 🔗https://github.com/rekkles2/Fed_WSVAD Document
-
🎧 Our inaugural Chinese-language audio drama Still Here is now streaming on Manbo. Creative Works
-
🥰 I’m honored to be invited as a presenter at the October Flower Monthly Event . Watch the talk ->>
-
🎉 One paper is accepted by IEEE Transactions on Industrial Informatics (Q1, JCR Top 4%, IF: 11.7)! Many thanks to my co-author and supervisor! Paper
-
🎉 One paper is accepted by Information Fusion (Q1, JCR Top 2%, IF: 15.5)! Many thanks to my co-author and supervisor! Paper
-
😎 A master’s student I mentored and collaborated with had a paper accepted to IJCNN 2025 — congratulations to my co-author! Paper
-
📚 Our Gaze-CIFAR10 Dataset is now available on GitHub (https://github.com/rekkles2/Gaze-CIFAR-10) Dataset
-
Selected Publications (view all )
Dual-Detector Reoptimization for Federated Weakly Supervised Video Anomaly Detection via Adaptive Dynamic Recursive Mapping
Dual-Detector Reoptimization for Federated Weakly Supervised Video Anomaly Detection via Adaptive Dynamic Recursive Mapping
Federated weakly-supervised video anomaly detection with mixture of local-to-global experts
Federated weakly-supervised video anomaly detection with mixture of local-to-global experts
All publications
Education
-
Sep. 2026 - Present -
Tianjin Normal University
B.Eng. in Intelligent Science and TechnologySep. 2022 - Jul. 2026
Experience
-
Feb. 2026 - Jul. 2026 -
JoyVoice Studio
Co-founder @JoyVoice StudioAug. 2025 - Present -
TJNU
Research Intern @Tianjin Key Laboratory of Wireless Mobile Communications and Power TransmissionAdvisor: Lect. Yong SuSep. 2023 - Jul. 2025
Honors & Awards
-
Tianjin Municipal Government Scholarship (CNY 8,000, Top <= 1%)Dec. 2025
-
Jiannanchun Science and Technology Innovation Scholarship (CNY 2,000, Top <= 1%)Nov. 2025
-
Contemporary Undergraduate Mathematical Contest in Modeling (National Second Prize, Top 2%)Nov. 2023