Zijie Song

Zijie Song

Ph.D. from Hefei University of Technology. His research interests include multimedia technology, cross-modal understanding, and vision-language learning.

Personal Information

Affiliation: School of Big Data and Statistics, Anhui University

Position: Lecturer

Degree: Ph.D.

Email: zjsong@ahu.edu.cn

Biography: Zijie Song is currently a Lecturer at the School of Big Data and Statistics, Anhui University. He received his Ph.D. degree in Computer Science and Technology from Hefei University of Technology in 2025 under the supervision of Prof. Richang Hong. His research interests include multimedia technology, cross-modal understanding, vision-language learning, and their applications.

He has participated in several research projects supported by the National Key Research and Development Program of China and the National Natural Science Foundation of China. He has published more than twenty papers in international journals and conferences, including TMM, NeurIPS, and ACM MM. He also serves as a reviewer for international journals and conferences, including TMM, TPAMI, and ACMMM. He is a member of the Multimedia Technical Committee of CSIG, the Editorial Board of Scientific Reports(SCI Q1).

News

  • The paper “B^3: Bridging Temporal Gap on Brown Bridge via Stochastic Process Based Modeling for Text-Video Retrieval” was published in IEEE Transactions on Multimedia, 2026. Zijie Song is the first author and corresponding author.

  • Zijie Song received the Invitation to join the Editorial Board of Scientific Reports (SCI Q1) and TPC Member for MMM 2027.

  • The paper “SDADL: SVD-Based Dual Adaptive Debiasing Learning for Personalized Recommendation” was published in IEEE Transactions on Computational Social Systems, 2026. Zijie Song is the corresponding author.

  • Zijie Song received the Best Poster Paper Award at ChinaMM 2026.

  • The paper “Image Captioning via Compact Bidirectional Architecture” was published in IEEE Transactions on Multimedia, 2026. Zijie Song is the first author.

  • The paper “Enhanced Keyframe Selection for Long-Video Question Answering via Large Language Model-Generated Pseudo-Labels” was published in Journal of Computer Research and Development, 2026, 63(6): 1563–1573. Zijie Song is the first author.

  • Zijie Song received the Best Poster Award at ChinaMM 2025.

Research Interests

Multimedia technology, cross-modal understanding, and vision-language learning.
Image captioning, video question answering, text-video retrieval, and image clustering.

Education

2022.08—2025.08Ph.D. in Computer Science and Technology, Hefei University of Technology
2019.09—2022.05M.Eng. in Electronics and Communication Engineering, Hefei University of Technology
2015.09—2019.06B.S. in Information and Computing Science, North China Electric Power University

Research Projects

2026.06—2027.06University–Industry Collaborative Research Project (Principal Investigator)
2022.01—2025.12General Program of the National Natural Science Foundation of China (Key Project Member)

Selected Publications

  1. Z. Song, L. Ge, Y. Wang, Y. Ma, Y. Zhou, Z. Hu, R. Hong, and M. Wang. B^3: Bridging Temporal Gap on Brown Bridge via Stochastic Process Based Modeling for Text-Video Retrieval. IEEE Transactions on Multimedia, 2026.

  2. Z. Song, Z. Hu, Y. Zhou, Y. Zhao, R. Hong, and M. Wang. Embedded Heterogeneous Attention Transformer for Cross-lingual Image Captioning. IEEE Transactions on Multimedia, 2024, 26: 9008–9020.

  3. Z. Song, Z. Hu, Y. Ma, J. Li, and R. Hong. Video Flow as Time Series: Discovering Temporal Consistency and Variability for VideoQA. IEEE International Conference on Multimedia and Expo (ICME), 2025.

  4. Z. Song, Y. Zhou, Z. Hu, D. Liu, H. Ben, R. Hong, and M. Wang. Image Captioning via Compact Bidirectional Architecture. IEEE Transactions on Multimedia, 2026.

  5. Z. Song, Z. Hu, and R. Hong. Efficient and Self-adaptive Rationale Knowledge Base for Visual Commonsense Reasoning. Multimedia Systems, 2023, 29(5): 3017–3026.

  6. Z. Song, Z. Hu, and R. Hong. Grid Jigsaw Representation with CLIP: a new perspective on image clustering. Multimedia Systems, 2025, 31(2), 105.

For more research work and academic service, please refer to:

Google Scholar: https://scholar.google.com/citations?user=99X3AYkAAAAJ
ORCID: https://orcid.org/0000-0002-1262-764X