Fan Xu

(Taken in April, 2025)
Esch-sur-Alzette
L-4364, Luxembourg

I’m currently a Ph.D. candidate in Multimodal Deep Learning under the supervision of Prof. Dr. Luis A. Leiva at University of Luxembourg. My research interests include Computer Vision, Large Language Models, Multimodal Models, Representation Learning, and Information Retrieval. I earned my Master’s degree in Mathematics in Data Science from Technical University of Munich (TUM) in Munich, Germany. Prior to that, I obtained my Bachelor’s degree in Mathematics in Financial Mathematics and Statistics from Northwest University (西北大学) in Xi’an, China.

Check my CV.

News

Oct. 23, 2026	Our work of Composed Image Retrieval on GLAM dataset has been accepted by Computational Humanities Research 2027. Congrats to all co-authors! 🎉
Aug. 21, 2026	Our work on Composed Image Retrieval using MLLMs and Diffusion models was accepted by Transactions on Machine Learning Research. Read the paper here.
Jun. 19, 2026	Our paper on aligning newspapers with posters was accepted by Digital Scholarship in the Humanities. Check here.

Selected Publications

Illustration of Composed Image Retrieval for Cultural Heritage Collections through Query Reformulation with Generative Multimodal Large Language Models

Composed Image Retrieval for Cultural Heritage Collections through Query Reformulation with Generative Multimodal Large Language Models

Fan Xu, Y. Mathansingh, and L. A. Leiva

In Computational Humanities Research Conference , Oct 2026

Code

@inproceedings{xu2026culturalheritage,
  author   = {Xu, Fan and Mathansingh, Yahanaa and Leiva, Luis A.},
  title    = {Composed Image Retrieval for Cultural Heritage Collections through Query Reformulation with Generative Multimodal Large Language Models},
  booktitle= {Computational Humanities Research Conference},
  year     = {2026},
  month    = {oct}
}

Training-Free Pseudo-Fusion for Composed Image Retrieval with Diffusion Models and Multimodal Large Language Models

Fan Xu, and L. A. Leiva

Transactions on Machine Learning Research, Aug 2026

HTML Code

@article{xu2026trainingfree,
  title   = {Training-Free Pseudo-Fusion for Composed Image Retrieval with Diffusion Models and Multimodal Large Language Models},
  author  = {Fan Xu and Luis A. Leiva},
  journal = {Transactions on Machine Learning Research},
  issn    = {2835-8856},
  year    = {2026},
  url     = {https://openreview.net/forum?id=6W3pFEQXZc}
}

Match Me if You Can: Aligning Historical Posters and Newspaper Articles via Multimodal Deep Learning

Fan Xu, M. Düring, and L. A. Leiva

Digital Scholarship in the Humanities, Jun 2026

HTML Code

@article{xu2026aligning,
  author  = {Xu, Fan and Düring, Marten and Leiva, Luis A},
  title   = {Match Me if You Can: Aligning Historical Posters and Newspaper articles via Multimodal Deep Learning},
  journal = {Digital Scholarship in the Humanities},
  pages   = {fqag088},
  year    = {2026},
  month   = {06},
  issn    = {2055-7671},
  doi     = {10.1093/llc/fqag088},
  url     = {https://doi.org/10.1093/llc/fqag088},
  eprint  = {https://academic.oup.com/dsh/advance-article-pdf/doi/10.1093/llc/fqag088/68561366/fqag088.pdf}
}