Congratulations to Savvina Daniil for winning the Women in RecSys Journal Paper of the Year award for her #IRRJ paper "On the challenges of studying bias in Recommender Systems"!
The award came with a free #RecSys2025 registration and the opportunity to present the IRRJ paper at the main conference.
With ACM RecSys25 right around the corner, we look forward to reconnecting with friends and old colleagues in Prague! This year, several Deezer researchers will be presenting their latest work: Tuesday – Paper Session 1 Bruno Sguerra will kick things off with "Biases in LLM-Generated Musical Taste Profiles for Recommendation", a study of fairness issues that arise when LLMs are used to generate user profiles. This work was carried out in collaboration with Elena E., Harin Lee, and Manuel M.. Tuesday - Posters Session 1 - "Beyond the Past: Leveraging Audio and Human Memory for Sequential Music Recommendation", by Viet-Anh TRAN, Bruno Sguerra, Gabriel Meseguer Brocal, Léa Briand, and Manuel M.. This work explores how to move beyond the limitations of past-only cognitive models by extrapolating music activation from audio similarity. 2 - "Just Ask for Music (JAM): Multimodal and Personalized Natural Language Music Recommendation", by Elena E., our friend Alessandro B. Melchiorre and his colleagues from JKU - Institute of Computational Perception. This work uses user preferences to translate query projections for more personalized music recommendation. Wednesday – Industry Symposium Léa Briand will join the session "The Perils of Production: Building Robust Learning Systems in the Wild." Friday - EARL workshop Clémence V. will present a poster on “Text2Playlist: Generating Personalized Playlists from Text on a Music Streaming Platform”, a framework for query-specific, personalized playlist generation deployed at Deezer! Work with Mathieu Delcluze and colleagues. Hope to catch up in Prague! ✨
IRRJ
Michael Ekstrand
Deezer Research
Hypertext 2026