"Does generative AI infringe copyright?" is an urgent question. It is al...
Neural language models are increasingly deployed into APIs and websites ...
We introduce MADLAD-400, a manually audited, general domain 3T token
mon...
Large language models are now tuned to align with the goals of their
cre...
Pretraining is the preliminary and fundamental step in developing capabl...
Model distillation is frequently proposed as a technique to reduce the
p...
Studying data memorization in neural language models helps us understand...
Large language models have been shown to achieve remarkable performance
...
Large language models (LMs) have been shown to memorize parts of their
t...
Natural language reflects our private lives and identities, making its
p...
Modern neural language models widely used in tasks across NLP risk memor...
We find that existing language modeling datasets contain many near-dupli...
It has become common to publish large (billion parameter) language model...
Neural networks have recently achieved human-level performance on variou...
Transfer learning, where a model is first pre-trained on a data-rich tas...