MorphNAS: Differentiable Architecture Search for Morphologically-Aware Multilingual NER
This work addresses a domain-specific problem for multilingual NLP, particularly in Indian languages, with an incremental approach by enhancing DARTS.
The paper tackles the challenge of processing morphologically complex languages, especially multiscript Indian languages, for Named Entity Recognition by introducing MorphNAS, a differentiable neural architecture search framework that incorporates linguistic meta-features, resulting in improved model proficiency.
Morphologically complex languages, particularly multiscript Indian languages, present significant challenges for Natural Language Processing (NLP). This work introduces MorphNAS, a novel differentiable neural architecture search framework designed to address these challenges. MorphNAS enhances Differentiable Architecture Search (DARTS) by incorporating linguistic meta-features such as script type and morphological complexity to optimize neural architectures for Named Entity Recognition (NER). It automatically identifies optimal micro-architectural elements tailored to language-specific morphology. By automating this search, MorphNAS aims to maximize the proficiency of multilingual NLP models, leading to improved comprehension and processing of these complex languages.