Interactive Refinement of Cross-Lingual Word Embeddings
This work addresses the challenge of improving classification accuracy in low-resource languages, offering an incremental refinement tool for NLP practitioners.
The paper tackles the problem of refining cross-lingual word embeddings for low-resource language classification by introducing CLIME, an interactive system that uses user annotations to update embeddings, resulting in higher test accuracy for health-related text identification in four languages.
Cross-lingual word embeddings transfer knowledge between languages: models trained on high-resource languages can predict in low-resource languages. We introduce CLIME, an interactive system to quickly refine cross-lingual word embeddings for a given classification problem. First, CLIME ranks words by their salience to the downstream task. Then, users mark similarity between keywords and their nearest neighbors in the embedding space. Finally, CLIME updates the embeddings using the annotations. We evaluate CLIME on identifying health-related text in four low-resource languages: Ilocano, Sinhalese, Tigrinya, and Uyghur. Embeddings refined by CLIME capture more nuanced word semantics and have higher test accuracy than the original embeddings. CLIME often improves accuracy faster than an active learning baseline and can be easily combined with active learning to improve results.