Maciej Stefaniak

According to our database, Maciej Stefaniak authored at least 4 papers in 2025.

Collaborative distances:

• Dijkstra number of four.
• Erdős number of three.

2025

μ-Parametrization for Mixture of Experts.

Jan Malasnicki, Kamil Ciebiera, Mateusz Borun, Maciej Pióro, Jan Ludziejewski, Maciej Stefaniak, Michal Krutul, Sebastian Jaszczur, Marek Cygan, Kamil Adamczewski, Jakub Krajewski

CoRR, August, 2025

Decoupled Relative Learning Rate Schedules.

Jan Ludziejewski, Jan Malasnicki, Maciej Pióro, Michal Krutul, Kamil Ciebiera, Maciej Stefaniak, Jakub Krajewski, Piotr Sankowski, Marek Cygan, Kamil Adamczewski, Sebastian Jaszczur

CoRR, July, 2025

Projected Compression: Trainable Projection for Efficient Transformer Compression.

Maciej Stefaniak, Michal Krutul, Jan Malasnicki, Maciej Pióro, Jakub Krajewski, Sebastian Jaszczur, Marek Cygan, Kamil Adamczewski, Jan Ludziejewski

CoRR, June, 2025

Joint MoE Scaling Laws: Mixture of Experts Can Be Memory Efficient.

Jan Ludziejewski, Maciej Pióro, Jakub Krajewski, Maciej Stefaniak, Michal Krutul, Jan Malasnicki, Marek Cygan, Piotr Sankowski, Kamil Adamczewski, Piotr Milos, Sebastian Jaszczur

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