Scholar
Guillaume Salha-Galvan
Google Scholar ID: BFT8u-oAAAAJ
Associate Professor, Shanghai Jiao Tong University
Machine Learning
Deep Learning
Graph Mining
Recommender Systems
Follow
Homepage
↗
Google Scholar
↗
Citations & Impact
All-time
Citations
851
H-index
14
i10-index
19
Publications
20
Co-authors
28
list available
Contact
Email
gsalhagalvan@gmail.com
CV
Open ↗
Twitter
Open ↗
GitHub
Open ↗
LinkedIn
Open ↗
Publications
6 items
IDProxy: Cold-Start CTR Prediction for Ads and Recommendation at Xiaohongshu with Multimodal LLMs
2026
Cited
0
MusicSem: A Semantically Rich Language--Audio Dataset of Natural Music Descriptions
2026
Cited
0
FairExpand: Individual Fairness on Graphs with Partial Similarity Information
2025
Cited
0
Exploring Large Action Sets with Hyperspherical Embeddings using von Mises-Fisher Sampling
2025
Cited
0
To Share or Not to Share: Investigating Weight Sharing in Variational Graph Autoencoders
2025
Cited
0
Contributions to Representation Learning with Graph Autoencoders and Applications to Music Recommendation
arXiv.org · 2022
Cited
3
Resume (English only)
Academic Achievements
Published over 30 peer-reviewed papers in top venues including ICML, KDD, WWW, IJCAI, EMNLP, and RecSys
Received best paper awards or honorable mentions at RecSys (2020, 2021, 2024) and ECIR (2024)
Research contributions have powered Deezer’s recommender systems, impacting millions of users
Papers accepted at ICML 2025 and WWW 2025
Co-organizing the EARL workshop on evaluating and applying recommender systems with LLMs at RecSys 2025
Co-authors
28 total
Romain Hennequin
Deezer Research
Michalis Vazirgiannis
Distinguished Prof, Ecole Polytechnique & Affiliate Prof. at MBZUAI
Walid Bendada
Deezer Research / Université Paris Dauphine
Co-author 4
Co-author 5
Manuel Moussallam
Deezer Research
Léa Briand
Deezer
Tristan Cazenave
Professor Université Paris Dauphine - PSL CNRS
×
Welcome back
Sign in to Agora
Welcome back! Please sign in to continue.
Email address
Password
Forgot password?
Continue
Do not have an account?
Sign up