Bilingual Lexicon Induction

Bilingual lexicon induction (BLI) aims to automatically discover word translations between languages without relying on pre-existing parallel corpora. Current research focuses on improving BLI performance, particularly for low-resource languages, through various approaches including matrix completion, graph matching, and leveraging the capabilities of large language models (LLMs) via prompting and fine-tuning, often incorporating contrastive learning or ranking methods. These advancements are significant because accurate bilingual lexicons are crucial for numerous natural language processing applications, such as machine translation and cross-lingual information retrieval, especially in scenarios with limited parallel data.