Piotr Nawrot
5 papers · 2022–2024 · 5 venues · across top CS/AI venues
Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.
Achievements
Jump to papers ↓
🌍
Conference Polyglot · Silver
🌉
Interdisciplinary Bridge · Bronze
🗺️
Taxonomy Completionist
(10)
Venues
Top co-authors
Research topics
Keywords
sequence modeling
(2)
transformer architecture
(1)
efficient training
(1)
autoregressive generation
(1)
language modeling
(1)
model pretraining
(1)
efficient computing
(1)
batch selection
(1)
dynamic pooling
(1)
learning rate decay
(1)
language model
(1)
compression ratio
(1)
model efficiency
(1)
throughput increase
(1)
hierarchical architecture
(1)
key-value cache compression
(1)
text-to-text transfer transformer
(1)
dynamic architecture
(1)
t5 model
(1)
transformer model
(1)
Papers
Dynamic Memory Compression: Retrofitting LLMs for Accelerated Inference
ICML 2024
Efficient Transformers with Dynamic Token Pooling
ACL 2023
nanoT5: Fast & Simple Pre-training and Fine-tuning of T5 Models with Limited Resources
EMNLP 2023
No Train No Gain: Revisiting Efficient Training Algorithms For Transformer-based Language Models
NeurIPS 2023
Hierarchical Transformers Are More Efficient Language Models
NAACL 2022