Simple knowledge distillation from mBERT produces smaller, faster Tagalog-only transformers that match the teacher on some tasks and lag on NER.
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
-
Extracting General-use Transformers for Low-resource Languages via Knowledge Distillation
Simple knowledge distillation from mBERT produces smaller, faster Tagalog-only transformers that match the teacher on some tasks and lag on NER.