Search papers, labs, and topics across Lattice.
This survey paper examines the role of transliteration in cross-lingual NLP, focusing on how it bridges the "script barrier" to improve transfer learning. It presents a taxonomy of motivations for using transliteration, including handling code-mixed text and leveraging language family relatedness, and analyzes different approaches for incorporating transliterations as input to language models. The paper highlights the trade-offs involved in different transliteration strategies and provides recommendations for researchers based on language, task, and resource constraints.
Transliteration, often overlooked, is a surprisingly versatile tool for boosting cross-lingual NLP, offering benefits from handling code-mixing to improving inference efficiency.
Cross-lingual transfer in NLP is often hindered by the ``script barrier''where differences in writing systems inhibit transfer learning between languages. Transliteration, the process of converting the script, has emerged as a powerful technique to bridge this gap by increasing lexical overlap. This paper provides a comprehensive survey of the application of transliteration in cross-lingual NLP. We present a taxonomy of key motivations to utilize transliterations in language models, and provide an overview of different approaches of incorporating transliterations as input. We analyze the evolution and effectiveness of these methods, discussing the critical trade-offs involved, and contextualize their need in modern LLMs. The review explores various settings that show how transliteration is beneficial, including handling code-mixed text, leveraging language family relatedness, and pragmatic gains in inference efficiency. Based on this analysis, we provide concrete recommendations for researchers on selecting and implementing the most appropriate transliteration strategy based on their specific language, task, and resource constraints.