publications

2026

1. Auditing Sybil: Explaining Deep Lung Cancer Risk Prediction Through Generative Interventional Attributions

   Bartłomiej Sobieski, Jakub Grzywaczewski, Karol Dobiczek, and 6 more authors

   In International Conference on Machine Learning (ICML), 2026
2. Local Intrinsic Dimension Unveils Hallucinations in Diffusion Models

   Bartłomiej Sobieski, Matthew Tivnan, D Płudowski, and 4 more authors

   arXiv preprint arXiv:2605.05026, 2026
3. Faithfulness Under the Distribution: A New Look at Attribution Evaluation

   Z Zhu, Z Jin, J Zhang, and 4 more authors

   In International Conference on Learning Representations (ICLR), 2026

2025

1. System-Embedded Diffusion Bridge Models

   Bartłomiej Sobieski, Matthew Tivnan, Yuang Wang, and 5 more authors

   In Advances in Neural Information Processing Systems (NeurIPS), 2025
2. Rethinking Visual Counterfactual Explanations Through Region Constraint

   Bartłomiej Sobieski, Jakub Grzywaczewski, Bartlomiej Sadlej, and 2 more authors

   In International Conference on Learning Representations (ICLR), 2025
3. Interpretable Machine Learning for Time-to-Event Prediction in Medicine and Healthcare

   Hubert Baniecki, Bartłomiej Sobieski, Patryk Szatkowski, and 2 more authors

   Artificial Intelligence in Medicine, 2025

2024

1. Global Counterfactual Directions

   Bartłomiej Sobieski and Przemyslaw Biecek

   In European Conference on Computer Vision (ECCV), 2024
2. Red-Teaming Segment Anything Model

   Krzysztof Jankowski, Bartłomiej Sobieski, Mateusz Kwiatkowski, and 4 more authors

   In Computer Vision and Pattern Recognition Conference Workshops (CVPRW), 2024

2023

1. Hospital Length of Stay Prediction Based on Multi-Modal Data Towards Trustworthy Human-AI Collaboration in Radiomics

   Hubert Baniecki, Bartłomiej Sobieski, Piotr Bombiński, and 2 more authors

   In International Conference on Artificial Intelligence in Medicine (AIME), 2023