Autoencoding is the best auxiliary objective for morphological inflection when unlabeled data is tiny; character-level masked language modeling wins with more unlabeled data, and segment oracle masking helps further.
When masking unlabeled data, we always sample 25% of tokens for masking
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
-
Improving Low-Resource Morphological Inflection via Self-Supervised Objectives
Autoencoding is the best auxiliary objective for morphological inflection when unlabeled data is tiny; character-level masked language modeling wins with more unlabeled data, and segment oracle masking helps further.