CV

For the most up-to-date version, please download the PDF (link on the right).

Table of contents

General Information

Full Name Marta Moscati
Languages Italian (native), English (C2), German (C1), French (B2), Chinese Mandarin (A1)

Education

  • 2021 - present
    PhD in Computer Science
    Johannes Kepler University Linz, Austria
    • Research on multimodal learning and recommender systems
    • Supervising students
    • Contributing to the scientific community as chair (publicity, proceedings), program committee, and reviewer
  • 2019
    PhD in Theoretical Particle Physics
    Karlsruhe Institute of Technology (KIT), Germany
    • Thesis "Lepton Flavour Non-Universality - From Effective Field Theory to Extended Gauge Models"
    • Research on new physics models and lepton flavor universality violation

Experience

  • 2025 - present
    Applied Scientist (L5)
    Albatross AI
    • Developing effective recommender systems for Albatross AI's customers
  • June 2024 - December 2024
    Research Intern
    Deezer Research Team, Paris, France
    • User modelling from large-scale music streaming data
    • Analysis of user behavior patterns in music consumption
  • 2021 - present
    PhD Researcher
    Johannes Kepler University Linz, Austria
    • Multimodal representation learning for recommender systems
    • Cold-start recommendation scenarios
    • Music information retrieval and emotion-based music recommendation
  • 2019 - 2021
    Postdoctoral Researcher
    Karlsruhe Institute of Technology (KIT), Germany
    • Research in theoretical particle physics
    • Focus on lepton flavor universality and beyond Standard Model physics

Selected Publications

  • 2026
    Face-Voice Association with Inductive Bias for Maximum Class Separation
    • M. Moscati, O. Kats, M. Noman, M.Z. Zaheer, Y. Hou, M. Schedl, S. Nawaz. IEEE ICASSP 2026.
  • 2025
    Parameter-Efficient Single Collaborative Branch for Recommendation
    • M. Moscati, S. Nawaz, M. Schedl. ACM RecSys 2025.
  • 2024
    A Multimodal Single-Branch Embedding Network for Recommendation in Cold-Start and Missing Modality Scenarios
    • C. Ganhör*, M. Moscati*, A. Hausberger, S. Nawaz, M. Schedl. ACM RecSys 2024. (*Equal contributions)
  • 2023
    Integrating the ACT-R Framework with Collaborative Filtering for Explainable Sequential Music Recommendation
    • M. Moscati, C. Wallmann, M. Reiter-Haas, D. Kowald, E. Lex, M. Schedl. ACM RecSys 2023.
  • 2022
    Music4All-Onion -- A Large-Scale Multi-Faceted Content-Centric Music Recommendation Dataset
    • M. Moscati, E. Parada-Cabaleiro, Y. Deldjoo, E. Zangerle, M. Schedl. ACM CIKM 2022.

Research Interests

  • Multimodal Learning
    • Cross-modal representation learning
    • Audio-visual learning (face-voice association)
    • Handling missing modalities in multimodal systems
  • Recommender Systems
    • Cold-start recommendation
    • Music recommendation
    • Sequential recommendation
  • Music Information Retrieval
    • Music emotion recognition
    • Multi-modal music retrieval (audio, lyrics, video)
    • Music discovery patterns and user behavior

Skills

Programming Languages
Machine Learning
Research

Other Interests

  • Passions: Music, Books, Mathematics, Science
  • Activities: Sports, Language learning