Michal Brzozowski

According to our database, Michal Brzozowski authored at least 5 papers between 2023 and 2026.

Collaborative distances:

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

2026

Reading the Finetuning Prior: Verbatim Content Recovery via Contrastive Decoding Diffing.

Michal Brzozowski, Zuzanna Dubanowska, Enrico Cassano, Neo Christopher Chung

CoRR, May, 2026

Aligned Training: A Parameter-Free Method to Improve Feature Quality and Stability of Sparse Autoencoders (SAE).

Michal Brzozowski, Neo Christopher Chung

CoRR, May, 2026

GPart: End-to-End Isometric Fine-Tuning via Global Parameter Partitioning.

Paolo Mandica, Michal Brzozowski, Zuzanna Dubanowska, Neo Christopher Chung

CoRR, May, 2026

2025

Representation-based Broad Hallucination Detectors Fail to Generalize Out of Distribution.

Zuzanna Dubanowska, Maciej Zelaszczyk, Michal Brzozowski, Paolo Mandica, Michal P. Karpowicz

Proceedings of the Findings of the Association for Computational Linguistics: EMNLP 2025, 2025

2023

Contrastive News and Social Media Linking using BERT for Articles and Tweets across Dual Platforms.

Jan Piotrowski, Marek Wachnicki, Mateusz Perlik, Jakub Podolak, Grzegorz Rucki, Michal Brzozowski, Pawel Olejnik, Julian Kozlowski, Tomasz Nocon, Jakub Koziel, Stanislaw Gizinski, Piotr Sankowski

CoRR, 2023