Filip Szatkowski

According to our database, Filip Szatkowski authored at least 17 papers between 2022 and 2026.

Collaborative distances:

• Dijkstra number of five.
• Erdős number of four.

2026

Beyond Classification: Dynamic Adapter Routing for Continual Multimodal Retrieval.

Alicja Dobrzeniecka, Filip Szatkowski, Sebastian Cygert, Szymon Lukasik, Bartlomiej Twardowski

CoRR, May, 2026

2025

Universal Properties of Activation Sparsity in Modern Large Language Models.

Filip Szatkowski, Patryk Bedkowski, Alessio Devoto, Jan Dubinski, Pasquale Minervini, Mikolaj Piórczynski, Simone Scardapane, Bartosz Wójcik

CoRR, September, 2025

Failure Prediction Is a Better Performance Proxy for Early-Exit Networks Than Calibration.

Piotr Kubaty, Filip Szatkowski, Metod Jazbec, Bartosz Wójcik

CoRR, August, 2025

Maybe I Should Not Answer That, but... Do LLMs Understand The Safety of Their Inputs?

Maciej Chrabaszcz, Filip Szatkowski, Bartosz Wójcik, Jan Dubinski, Tomasz Trzcinski

CoRR, February, 2025

Exploring the Stability Gap in Continual Learning: The Role of the Classification Head.

Wojciech Lapacz, Daniel Marczak, Filip Szatkowski, Tomasz Trzcinski

Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 2025

Improving Continual Learning Performance and Efficiency with Auxiliary Classifiers.

Filip Szatkowski, Yaoyue Zheng, Fei Yang, Tomasz Trzcinski, Bartlomiej Twardowski, Joost van de Weijer

Proceedings of the Forty-second International Conference on Machine Learning, 2025

ExpertSim: Fast Particle Detector Simulation Using Mixture-of-Generative-Experts.

Patryk Bedkowski, Jan Dubinski, Filip Szatkowski, Kamil Deja, Przemyslaw Rokita, Tomasz Trzcinski

Proceedings of the ECAI 2025 - 28th European Conference on Artificial Intelligence, 25-30 October 2025, Bologna, Italy, 2025

2024

Sparser, Better, Deeper, Stronger: Improving Sparse Training with Exact Orthogonal Initialization.

Aleksandra Irena Nowak, Lukasz Gniecki, Filip Szatkowski, Jacek Tabor

CoRR, 2024

Accelerated Inference and Reduced Forgetting: The Dual Benefits of Early-Exit Networks in Continual Learning.

Filip Szatkowski, Fei Yang, Bartlomiej Twardowski, Tomasz Trzcinski, Joost van de Weijer

CoRR, 2024

Adapt Your Teacher: Improving Knowledge Distillation for Exemplar-free Continual Learning.

Filip Szatkowski, Mateusz Pyla, Marcin Przewiezlikowski, Sebastian Cygert, Bartlomiej Twardowski, Tomasz Trzcinski

Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 2024

Exploiting Activation Sparsity with Dense to Dynamic-k Mixture-of-Experts Conversion.

Filip Szatkowski, Bartosz Wójcik, Mikolaj Piórczynski, Simone Scardapane

Proceedings of the Advances in Neural Information Processing Systems 37: Annual Conference on Neural Information Processing Systems 2024, 2024

Sparser, Better, Deeper, Stronger: Improving Static Sparse Training with Exact Orthogonal Initialization.

Aleksandra Nowak, Lukasz Gniecki, Filip Szatkowski, Jacek Tabor

Proceedings of the Forty-first International Conference on Machine Learning, 2024

2023

Zero time waste in pre-trained early exit neural networks.

Bartosz Wójcik, Marcin Przewiezlikowski, Filip Szatkowski, Maciej Wolczyk, Klaudia Balazy, Bartlomiej Krzepkowski, Igor T. Podolak, Jacek Tabor, Marek Smieja, Tomasz Trzcinski

Neural Networks, November, 2023

Exploiting Transformer Activation Sparsity with Dynamic Inference.

Mikolaj Piórczynski, Filip Szatkowski, Klaudia Balazy, Bartosz Wójcik

CoRR, 2023

Hypernetworks Build Implicit Neural Representations of Sounds.

Filip Szatkowski, Karol J. Piczak, Przemyslaw Spurek, Jacek Tabor, Tomasz Trzcinski

Proceedings of the Machine Learning and Knowledge Discovery in Databases: Research Track, 2023

2022

HyperSound: Generating Implicit Neural Representations of Audio Signals with Hypernetworks.

Filip Szatkowski, Karol J. Piczak, Przemyslaw Spurek, Jacek Tabor, Tomasz Trzcinski

CoRR, 2022

Progressive Latent Replay for Efficient Generative Rehearsal.

Stanislaw Pawlak, Filip Szatkowski, Michal Bortkiewicz, Jan Dubinski, Tomasz Trzcinski

Proceedings of the Neural Information Processing - 29th International Conference, 2022