Using 16-bit mixed precision with distributed data parallelism speeds up BERT domain adaptation up to 3.5 times while keeping validation accuracy similar.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Low-resource domain adaptation while minimizing energy and hardware resource consumption
Using 16-bit mixed precision with distributed data parallelism speeds up BERT domain adaptation up to 3.5 times while keeping validation accuracy similar.