Luca Marinelli
Postdoctoral Researcher in Computational Social Science @ Weizenbaum Institute.
Luca is a Postdoctoral Researcher in the GENDIS project at the Weizenbaum Institute, studying gender-based disinformation on online platforms with a focus on visual and multimodal data analysis. The project aims to develop computational methods and an open-source detection tool with academic and practice partners in Berlin and Munich.
Luca holds a PhD in Artificial Intelligence and Music from Queen Mary University of London. Grounded in content and discourse analysis, his doctoral work combined computational and critical approaches to study how gender stereotypes are encoded across multiple modalities in media (with a focus on music), using various machine learning techniques, LLMs (retrieval-augmented generation), and explainable AI.
news
| Oct 01, 2026 | Today I officially started a Postdoc position in Berlin at the Weizenbaum Institute!!! |
|---|---|
| Apr 10, 2026 | The thesis is available online at https://qmro.qmul.ac.uk/items/0cd81acb-67c9-47f4-ba8a-cf78362e8969. |
| Mar 16, 2026 | PhD viva passed without corrections! The thesis will be published in the next few months. |
| Jan 05, 2026 | I submitted the thesis! The discussion will be sometime in March. |
| Nov 20, 2025 | Invited lecture on “Gender‑Coded Sound: Computational Insights into Toy Advertising Music” at the Hochschule für Musik, Theater und Medien Hannover, Germany. |
selected publications
- Leveraging RAG for a Low-Resource Audio-Aware Diachronic Analysis of Gendered Toy MarketingIn Proceedings of the First Workshop on Natural Language Processing and Language Models for Digital Humanities @ RANLP 2025, 2025
- Explainable Modeling of Gender-Targeting Practices in Toy Advertising Sound and MusicIn 2024 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW), 2024
- Gender-Coded Sound: Analysing the Gendering of Music in Toy Commercials via Multi-Task Learning24th International Society for Music Information Retrieval Conference (ISMIR 2023), 2023
- Musical Dynamics Classification with CNN and Modulation SpectraIn Proceedings of the 17th Sound and Music Computing Conference, Torino, Italy, 2020