Seguir
Piotr Nawrot
Piotr Nawrot
University of Edinburgh
E-mail confirmado em ed.ac.uk - Página inicial
Título
Citado por
Citado por
Ano
Hierarchical transformers are more efficient language models
P Nawrot, S Tworkowski, M Tyrolski, Ł Kaiser, Y Wu, C Szegedy, ...
arXiv preprint arXiv:2110.13711, 2021
332021
No train no gain: Revisiting efficient training algorithms for transformer-based language models
J Kaddour, O Key, P Nawrot, P Minervini, MJ Kusner
Advances in Neural Information Processing Systems 36, 2024
142024
Efficient transformers with dynamic token pooling
P Nawrot, J Chorowski, A Łańcucki, EM Ponti
arXiv preprint arXiv:2211.09761, 2022
122022
Dynamic Memory Compression: Retrofitting LLMs for Accelerated Inference
P Nawrot, A Łańcucki, M Chochowski, D Tarjan, EM Ponti
arXiv preprint arXiv:2403.09636, 2024
22024
nanot5: A pytorch framework for pre-training and fine-tuning t5-style models with limited resources
P Nawrot
arXiv preprint arXiv:2309.02373, 2023
2*2023
O sistema não pode executar a operação agora. Tente novamente mais tarde.
Artigos 1–5