• I’m Yuxin Lu (陆宇昕), an undergraduate student majoring in Data Science and Big Data Technology at the School of Future Science and Engineering, Soochow University.
  • In recent years, I have been focusing on generative models. Under the guidance of Prof. Min Cao, Qian Qiao and Jia Gong, I have conducted in-depth research in the fields of image generation and video generation.
  • My long-term research interests are focused on Video Generation, Embodied AI, and Multimodal Large Language Models.
  • I am currently preparing for future Master’s/Ph.D. applications and looking forward to further diving into Artificial Intelligence. You can view my CV or contact me at junesnow.lyx@gmail.com.

🔥 News

  • 2026.05:  📄 AsymTalker is under review at NeurIPS 2026.
  • 2026.03:  🔬 Research internship at Shanghai Academy of AI for Science.
  • 2026.03:  📄 SynTBPR is under Major Revision at IEEE TIFS.
  • 2025.07:  🔬 Research internship at Wuhan Institute of Artificial Intelligence.
  • 2024.08:  🌐 Personal blog launched at junesnow.top.

📝 Research Publication

  • An Empirical Study of Validating Synthetic Data for Text-Based Person Retrieval

Min Cao, Yuxin Lu, Ziyin Zeng, Mang Ye, Dong Yi, Jinqiao Wang

IEEE TIFS, Major Revision; CCF-A, SCI Q1

Paper

  • AsymTalker: Identity-Consistent Long-Term Talking Head Generation via Asymmetric Distillation

Yuxin Lu*, Qian Qiao*, Jiayang Sun*, Guibo Zhu, Min Cao

NeurIPS 2026, Under Review; CCF-A

Paper

📖 Education & Experience

B.Sc. in Data Science, Soochow University
Sep 2023 – Present
Soochow University logo
Research Intern, Shanghai Academy of AI for Science
Mar 2026 – Apr 2026
Shanghai Academy of AI for Science logo
Research Intern, Wuhan Institute of Artificial Intelligence
Jul 2025 – Apr 2026
Wuhan Institute of Artificial Intelligence logo

🔬 Research Interests

  • Video Generation
  • Embodied AI
  • Multimodal Large Language Models

🧩 Projects

SynTBPR

SynTBPR

📝 An Empirical Study of Validating Synthetic Data for Text-Based Person Retrieval

🔗 Project

  • This project studies the validity of synthetic data for text-based person retrieval.
  • The work explores how synthetic data can support retrieval model training and evaluation under practical data constraints.
  • The project is implemented in Python and is publicly available on GitHub.

✍️ Blog & Notes

📬 Contact

Feel free to contact me at junesnow.lyx@gmail.com to discuss research ideas, collaborations, or academic opportunities.