Kun Wang

Ph.D. Candidate in Software Engineering

Intelligent Media Research Center (iLearn)

My research focuses on multimedia retrieval in real-world scenarios, with particular interest in fine-grained video understanding, cross-modal semantic alignment, and efficient multimodal representation.

News

Apr 2026Our ACM TOIS paper has been invited for presentation at SIGIR 2026.
Apr 2026One paper has been accepted by SIGIR 2026.
Mar 2026Our team won 1st place in the NTIRE 2026 Video Saliency Prediction Challenge at CVPR 2026.
Mar 2026One paper has been accepted by CVPR 2026.
Dec 2025One paper has been accepted by IEEE TCSVT.
Dec 2025One paper has been accepted by IEEE TMM.
Nov 2025One paper has been accepted by ACM TOIS.
Mar 2025One paper has been accepted by SIGIR 2025.

Selected Publications

Redundancy Mitigation: Towards Accurate and Efficient Image-Text Retrieval
Kun Wang, Yupeng Hu, Hao Liu, Lirong Jie, Liqiang Nie
IEEE TCSVT
Cross-Modal Representation Shift Refinement for Point-supervised Video Moment Retrieval
Kun Wang, Yupeng Hu, Hao Liu, Jiang Shao, Liqiang Nie
ACM TOIS
Explicit Granularity and Implicit Scale Correspondence Learning for Point-Supervised Video Moment Localization
Kun Wang, Hao Liu, Lirong Jie, Zixu Li, Yupeng Hu, Liqiang Nie
ACM MM