Arabizi vs LLMs: Can the Genie Understand the Language of Aladdin?

TL;DR

This paper evaluates the ability of various large language models to translate Arabizi, a hybrid Arabic dialect, into Modern Standard Arabic and English, highlighting challenges and dialect-specific performance issues.

Contribution

It provides a comprehensive analysis of LLMs' effectiveness in translating Arabizi, a less-studied dialect, into multiple languages, using both human and automatic evaluation methods.

Findings

LLMs show varying performance across dialects

Translations into English often outperform those into Arabic

Certain dialects are more accurately translated than others

Abstract

In this era of rapid technological advancements, communication continues to evolve as new linguistic phenomena emerge. Among these is Arabizi, a hybrid form of Arabic that incorporates Latin characters and numbers to represent the spoken dialects of Arab communities. Arabizi is widely used on social media and allows people to communicate in an informal and dynamic way, but it poses significant challenges for machine translation due to its lack of formal structure and deeply embedded cultural nuances. This case study arises from a growing need to translate Arabizi for gisting purposes. It evaluates the capacity of different LLMs to decode and translate Arabizi, focusing on multiple Arabic dialects that have rarely been studied up until now. Using a combination of human evaluators and automatic metrics, this research project investigates the model's performance in translating Arabizi into both Modern Standard Arabic and English. Key questions explored include which dialects are translated most effectively and whether translations into English surpass those into Arabic.