Pierre Lubitzsch

Pierre Lubitzsch

Ph.D. Student at BIFOLD & TU Berlin

Publications

You can also find my articles on my Google Scholar profile.

Conference Papers

ERASE - A Real-World Aligned Benchmark for Unlearning in Recommender Systems

Published in SIGIR '26, 2026

We present ERASE, a large-scale benchmark for machine unlearning in recommender systems designed to align with real-world usage, spanning collaborative filtering, session-based, and next-basket recommendation, covering seven unlearning algorithms across nine datasets and nine state-of-the-art models.

Recommended citation: Pierre Lubitzsch, Maarten de Rijke, Sebastian Schelter. (2026). "ERASE - A Real-World Aligned Benchmark for Unlearning in Recommender Systems." In Proceedings of the 49th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR '26), pp. 3310–3318.
Download Paper

Towards a Real-World Aligned Benchmark for Unlearning in Recommender Systems

Published in FAccTRec@RecSys 2025, 2025

We propose design desiderata and research questions to guide the development of a realistic benchmark for machine unlearning in recommender systems, spanning multiple recommendation tasks and unlearning scenarios.

Recommended citation: Pierre Lubitzsch, Olga Ovcharenko, Hao Chen, Maarten de Rijke, Sebastian Schelter. (2025). "Towards a Real-World Aligned Benchmark for Unlearning in Recommender Systems." FAccTRec@RecSys 2025
Download Paper