Multilingual Large Language Models Are Not (Yet) Code-Switchers

TL;DR

This paper empirically evaluates multilingual large language models' ability to handle code-switching, revealing they underperform compared to smaller fine-tuned models and highlighting the need for targeted research in this area.

Contribution

It provides a comprehensive benchmark of multilingual LLMs on code-switching tasks, demonstrating their limitations and the gap between multilingual capabilities and code-switching proficiency.

Findings

Multilingual LLMs perform variably across tasks with zero/few-shot prompting.

They underperform compared to smaller fine-tuned models in code-switching tasks.

Current multilingualism in LLMs does not ensure effective handling of code-switching texts.

Abstract

Multilingual Large Language Models (LLMs) have recently shown great capabilities in a wide range of tasks, exhibiting state-of-the-art performance through zero-shot or few-shot prompting methods. While there have been extensive studies on their abilities in monolingual tasks, the investigation of their potential in the context of code-switching (CSW), the practice of alternating languages within an utterance, remains relatively uncharted. In this paper, we provide a comprehensive empirical analysis of various multilingual LLMs, benchmarking their performance across four tasks: sentiment analysis, machine translation, summarization and word-level language identification. Our results indicate that despite multilingual LLMs exhibiting promising outcomes in certain tasks using zero or few-shot prompting, they still underperform in comparison to fine-tuned models of much smaller scales. We argue that current "multilingualism" in LLMs does not inherently imply proficiency with code-switching texts, calling for future research to bridge this discrepancy.