Magdalena Proszewska

I am currently a PhD student in the ELLIS program at the University of Edinburgh, supervised by Siddharth Narayanaswamy and co-supervised by Stephan Günnemann from TU Münich.

My research focuses on improving neural networks for graph data, and diffusion models.

I have completed my master’s studies in Mathematics and Computer Science at Jagiellonian University, where I collaborated with the Group of Machine Learning Research GMUM.

Outside of academia, I completed a few ML internships at Google and Amazon.

selected publications

  1. On Designing Diffusion Autoencoders for Efficient Generation and Representation Learning
    Magdalena Proszewska, Nikolay Malkin, and N. Siddharth
    arXiv preprint arXiv:2506.00136, May 2025
    HTML
  2. B-XAIC Dataset: Benchmarking Explainable AI for Graph Neural Networks Using Chemical Data
    Magdalena Proszewska, Tomasz Danel, and Dawid Rymarczyk
    arXiv preprint arXiv:2505.22252, May 2025
    HTML
  3. Bernoulli Priors as Efficient Denoising Guides for Diffusion Models
    Magdalena Proszewska, Nikolay Malkin, and N. Siddharth
    3rd Workshop on Generative Models for Computer Vision at the IEEE/CVF Conference on Computer Vision and Pattern Recognition 2025 (CVPR), Jun 2025
    HTML
  4. Multi-Label Conditional Generation From Pre-Trained Models
    Magdalena Proszewska, Maciej Wołczyk, Maciej Zieba, and 3 more authors
    IEEE Transactions on Pattern Analysis and Machine Intelligence, Aug 2024
    HTML
  5. GlowVC: Mel-spectrogram space disentangling model for language-independent text-free voice conversion
    Magdalena Proszewska, Grzegorz Beringer, Daniel Sáez-Trigueros, and 3 more authors
    In Interspeech 2022 , Sep 2022
    HTML