Nawrot, Piotr

3 publications

NeurIPS 2025 Inference-Time Hyper-Scaling with KV Cache Compression Adrian Łańcucki, Konrad Staniszewski, Piotr Nawrot, Edoardo Ponti

ICML 2024 Dynamic Memory Compression: Retrofitting LLMs for Accelerated Inference Piotr Nawrot, Adrian Łańcucki, Marcin Chochowski, David Tarjan, Edoardo Ponti

NeurIPS 2023 No Train No Gain: Revisiting Efficient Training Algorithms for Transformer-Based Language Models Jean Kaddour, Oscar Key, Piotr Nawrot, Pasquale Minervini, Matt J Kusner