pith. sign in

arxiv: 2401.14280 · v3 · pith:TJROXMENnew · submitted 2024-01-25 · 💻 cs.CL · cs.AI

RomanSetu: Efficiently unlocking multilingual capabilities of Large Language Models via Romanization

classification 💻 cs.CL cs.AI
keywords romanizedenglishtextapproachlanguagesllmsscriptalignment
0
0 comments X
read the original abstract

This study addresses the challenge of extending Large Language Models (LLMs) to non-English languages that use non-Roman scripts. We propose an approach that utilizes the romanized form of text as an interface for LLMs, hypothesizing that its frequent informal use and shared tokens with English enhance cross-lingual alignment. Our approach involves the continual pretraining of an English LLM like Llama 2 on romanized text of non-English, non-Roman script languages, followed by instruction tuning on romanized data. The results indicate that romanized text not only reduces token fertility by 2x-4x but also matches or outperforms native script representation across various NLU, NLG, and MT tasks. Moreover, the embeddings computed on romanized text exhibit closer alignment with their English translations than those from the native script. Our approach presents a promising direction for leveraging the power of English LLMs in languages traditionally underrepresented in NLP. Our code is available on https://github.com/AI4Bharat/romansetu.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.