Filip Szatkowski

22 papers A* 3A 3Journal 13Unranked 3

Year	Rank	Type	Title / Venue / Authors
2025	A	`conf`	ExpertSim: Fast Particle Detector Simulation Using Mixture-of-Generative-Experts. ECAI Patryk Bedkowski, Jan Dubinski, Filip Szatkowski, Kamil Deja, Przemyslaw Rokita, Tomasz Trzcinski
2025	J	`jnl`	ExpertSim: Fast Particle Detector Simulation Using Mixture-of-Generative-Experts. CoRR Patryk Bedkowski, Jan Dubinski, Filip Szatkowski, Kamil Deja, Przemyslaw Rokita, Tomasz Trzcinski
2025	A	`conf`	Exploring the Stability Gap in Continual Learning: The Role of the Classification Head. WACV Wojciech Lapacz, Daniel Marczak, Filip Szatkowski, Tomasz Trzcinski
2025	J	`jnl`	Failure Prediction Is a Better Performance Proxy for Early-Exit Networks Than Calibration. CoRR Piotr Kubaty, Filip Szatkowski, Metod Jazbec, Bartosz Wójcik
2025	A*	`conf`	Improving Continual Learning Performance and Efficiency with Auxiliary Classifiers. ICML Filip Szatkowski, Yaoyue Zheng, Fei Yang, Tomasz Trzcinski, Bartlomiej Twardowski, Joost van de Weijer
2025	J	`jnl`	Maybe I Should Not Answer That, but... Do LLMs Understand The Safety of Their Inputs? CoRR Maciej Chrabaszcz, Filip Szatkowski, Bartosz Wójcik, Jan Dubinski, Tomasz Trzcinski
2025	J	`jnl`	Universal Properties of Activation Sparsity in Modern Large Language Models. CoRR Filip Szatkowski, Patryk Bedkowski, Alessio Devoto, Jan Dubinski, Pasquale Minervini, Mikolaj Piórczynski, Simone Scardapane, Bartosz Wójcik
2024	J	`jnl`	Accelerated Inference and Reduced Forgetting: The Dual Benefits of Early-Exit Networks in Continual Learning. CoRR Filip Szatkowski, Fei Yang, Bartlomiej Twardowski, Tomasz Trzcinski, Joost van de Weijer
2024	A	`conf`	Adapt Your Teacher: Improving Knowledge Distillation for Exemplar-free Continual Learning. WACV Filip Szatkowski, Mateusz Pyla, Marcin Przewiezlikowski, Sebastian Cygert, Bartlomiej Twardowski, Tomasz Trzcinski
2024	A*	`conf`	Exploiting Activation Sparsity with Dense to Dynamic-k Mixture-of-Experts Conversion. NeurIPS Filip Szatkowski, Bartosz Wójcik, Mikolaj Piórczynski, Simone Scardapane
2024	J	`jnl`	Exploring the Stability Gap in Continual Learning: The Role of the Classification Head. CoRR Wojciech Lapacz, Daniel Marczak, Filip Szatkowski, Tomasz Trzcinski
2024	J	`jnl`	Sparser, Better, Deeper, Stronger: Improving Sparse Training with Exact Orthogonal Initialization. CoRR Aleksandra Irena Nowak, Lukasz Gniecki, Filip Szatkowski, Jacek Tabor
2024	A*	`conf`	Sparser, Better, Deeper, Stronger: Improving Static Sparse Training with Exact Orthogonal Initialization. ICML Aleksandra Nowak, Lukasz Gniecki, Filip Szatkowski, Jacek Tabor
2023	—	`conf`	Adapt Your Teacher: Improving Knowledge Distillation for Exemplar-free Continual Learning. ICCV (Workshops) Filip Szatkowski, Mateusz Pyla, Marcin Przewiezlikowski, Sebastian Cygert, Bartlomiej Twardowski, Tomasz Trzcinski
2023	J	`jnl`	Adapt Your Teacher: Improving Knowledge Distillation for Exemplar-free Continual Learning. CoRR Filip Szatkowski, Mateusz Pyla, Marcin Przewiezlikowski, Sebastian Cygert, Bartlomiej Twardowski, Tomasz Trzcinski
2023	J	`jnl`	Exploiting Transformer Activation Sparsity with Dynamic Inference. CoRR Mikolaj Piórczynski, Filip Szatkowski, Klaudia Balazy, Bartosz Wójcik
2023	—	`conf`	Hypernetworks Build Implicit Neural Representations of Sounds. ECML/PKDD (4) Filip Szatkowski, Karol J. Piczak, Przemyslaw Spurek, Jacek Tabor, Tomasz Trzcinski
2023	J	`jnl`	Hypernetworks build Implicit Neural Representations of Sounds. CoRR Filip Szatkowski, Karol J. Piczak, Przemyslaw Spurek, Jacek Tabor, Tomasz Trzcinski
2023	J	`jnl`	Zero time waste in pre-trained early exit neural networks. Neural Networks Bartosz Wójcik, Marcin Przewiezlikowski, Filip Szatkowski, Maciej Wolczyk, Klaudia Balazy, Bartlomiej Krzepkowski, Igor T. Podolak, Jacek Tabor, Marek Smieja, Tomasz Trzcinski
2022	J	`jnl`	HyperSound: Generating Implicit Neural Representations of Audio Signals with Hypernetworks. CoRR Filip Szatkowski, Karol J. Piczak, Przemyslaw Spurek, Jacek Tabor, Tomasz Trzcinski
2022	—	`conf`	Progressive Latent Replay for Efficient Generative Rehearsal. ICONIP (4) Stanislaw Pawlak, Filip Szatkowski, Michal Bortkiewicz, Jan Dubinski, Tomasz Trzcinski
2022	J	`jnl`	Progressive Latent Replay for efficient Generative Rehearsal. CoRR Stanislaw Pawlak, Filip Szatkowski, Michal Bortkiewicz, Jan Dubinski, Tomasz Trzcinski