Morphologically Aware Word-Level Translation
This work addresses the issue of ignoring inflectional morphology in bilingual lexicon induction, which is a problem for NLP researchers and practitioners, and is incremental as it builds on existing methods with specific enhancements.
The paper tackles the problem of bilingual lexicon induction by proposing a morphologically aware probability model that jointly models lexeme translation and inflectional morphology, resulting in a 19% average accuracy improvement in supervised settings and 16% in weakly supervised settings across 6 language pairs.
We propose a novel morphologically aware probability model for bilingual lexicon induction, which jointly models lexeme translation and inflectional morphology in a structured way. Our model exploits the basic linguistic intuition that the lexeme is the key lexical unit of meaning, while inflectional morphology provides additional syntactic information. This approach leads to substantial performance improvements - 19% average improvement in accuracy across 6 language pairs over the state of the art in the supervised setting and 16% in the weakly supervised setting. As another contribution, we highlight issues associated with modern BLI that stem from ignoring inflectional morphology, and propose three suggestions for improving the task.