LLMs Are Few-Shot In-Context Low-Resource Language Learners

TL;DR

This paper investigates the effectiveness of in-context learning with large language models for low-resource languages, proposing a new query alignment method to improve semantic understanding and cross-lingual transfer.

Contribution

It provides a comprehensive study of ICL in low-resource languages, introduces query alignment as an alternative to label alignment, and offers insights into enhancing LLM performance in underrepresented languages.

Findings

ICL significantly improves low-resource language understanding.

Query alignment outperforms label alignment in cross-lingual tasks.

Semantic relevance in in-context examples enhances model performance.

Abstract

In-context learning (ICL) empowers large language models (LLMs) to perform diverse tasks in underrepresented languages using only short in-context information, offering a crucial avenue for narrowing the gap between high-resource and low-resource languages. Nonetheless, there is only a handful of works explored ICL for low-resource languages with most of them focusing on relatively high-resource languages, such as French and Spanish. In this work, we extensively study ICL and its cross-lingual variation (X-ICL) on 25 low-resource and 7 relatively higher-resource languages. Our study not only assesses the effectiveness of ICL with LLMs in low-resource languages but also identifies the shortcomings of in-context label alignment, and introduces a more effective alternative: query alignment. Moreover, we provide valuable insights into various facets of ICL for low-resource languages. Our study concludes the significance of few-shot in-context information on enhancing the low-resource understanding quality of LLMs through semantically relevant information by closing the language gap in the target language and aligning the semantics between the targeted low-resource and the high-resource language that the model is proficient in. Our work highlights the importance of advancing ICL research, particularly for low-resource languages. Our code is publicly released at https://github.com/SamuelCahyawijaya/in-context-alignment