CWoMP: Morpheme Representation Learning for Interlinear Glossing
This addresses the laborious task of IGT production for linguists, offering an efficient and interpretable solution, though it is incremental as it builds on prior automated methods by incorporating morpheme-level processing.
The paper tackles the problem of automating interlinear glossed text (IGT) generation for language documentation by proposing CWoMP, a method that learns morpheme representations to handle compositional structure, resulting in outperforming existing methods with strong gains in low-resource settings.
Interlinear glossed text (IGT) is a standard notation for language documentation which is linguistically rich but laborious to produce manually. Recent automated IGT methods treat glosses as character sequences, neglecting their compositional structure. We propose CWoMP (Contrastive Word-Morpheme Pretraining), which instead treats morphemes as atomic form-meaning units with learned representations. A contrastively trained encoder aligns words-in-context with their constituent morphemes in a shared embedding space; an autoregressive decoder then generates the morpheme sequence by retrieving entries from a mutable lexicon of these embeddings. Predictions are interpretable--grounded in lexicon entries--and users can improve results at inference time by expanding the lexicon without retraining. We evaluate on diverse low-resource languages, showing that CWoMP outperforms existing methods while being significantly more efficient, with particularly strong gains in extremely low-resource settings.