Associated with the Linz Institute of Technology and the Institute of Computational Perception @ JKU Linz, Austria

Single-Branch Architectures for Recommendation

daquamrec_single Teaser

Abstract

Single-branch architectures have been proven effective for several multimodal learning tasks. In this talk, after reviewing the use of single-branch architectures in multimodal learning, I describe their use in multimodal recommendation, showing how they allow to address missing-modality and cold-start scenarios. I then describe the use of single-branch architectures in collaborative filtering, showing how they allow to reduce the number of model parameters without substantially affecting the quality of recommendations.


Citation

Marta Moscati
Single-Branch Architectures for Recommendation
Proceedings of the DaQuaMRec 2025 Workshop on Data Quality-Aware Multimodal Recommendation co-located with RecSys 2025, Prague, Czech Republic., 4188: 2025.

BibTeX

@proceedings{Moscati2025daquamrec_single,
    title = {Single-Branch Architectures for Recommendation},
    author = {Moscati, Marta},
    booktitle = {Proceedings of the DaQuaMRec 2025 Workshop on Data Quality-Aware Multimodal Recommendation co-located with RecSys 2025, Prague, Czech Republic.},
    editor = {Pomo, Claudio and Jannach, Dietmar and Kim, Yubin and Malitesta, Daniele and Mancino, Alberto Carlo Maria and McAuley, Julian and Melchiorre, Alessandro and Nawaz, Shah},
    publisher = {CEUR-WS.org},
    url = {https://ceur-ws.org/Vol-4188/},
    volume = {4188},
    year = {2025}
}