PhD student, Jagiellonian University in Krakow
2 papers at NeurIPS 2025
We introduce ZEUS, a transformer pretrained on synthetic datasets to perform zero‐shot, fully unsupervised clustering of tabular data—no fine‐tuning or hyperparameter search needed—and show it matches or outperforms classical and deep baselines.
DiCoFlex generates diverse, customizable counterfactual explanations in a single forward pass without requiring model access, enabling real-time constraint adjustments
