Left Behind: Cross-Lingual Transfer as a Bridge for Low-Resource Languages in Large Language Models

Domaine:

natural language processing

Type de record:

paperdataset

Créateur:

Bei

Hôte:

We investigate how large language models perform on low-resource languages by benchmarking eight LLMs across five experimental conditions in English, Kazakh, and Mongolian. Using 50 hand-crafted questions spanning factual, reasoning, technical, and culturally grounded categories, we evaluate 2,000 responses on accuracy, fluency, and completeness. We find a consistent performance gap of 13.8-16.7 percentage points between English and low-resource language conditions, with models maintaining surface-level fluency while producing significantly less accurate content. Cross-lingual transfer-prompting models to reason in English before translating back-yields selective gains for bilingual architectures (+2.2pp to +4.3pp) but provides no benefit to English-dominant models. Our results demonstrate that current LLMs systematically underserve low-resource language communities, and that effective mitigation strategies are architecture-dependent rather than universal.

Visit

arxiv.org

Similaires

Few-Shot Cross-Lingual Transfer for Prompting Large Language Models in Low-Resource LanguagesZero-Shot Cross-Lingual Reranking with Large Language Models for Low-Resource LanguagesCross-lingual NER Transfer with Pretrained Language Models: Accuracy Degradation in Low-Resource LanguagesCross-lingual transfer of multilingual models on low resource African Languageskaranwxliaa/Cross-lingual-transfer-of-multilingual-models-on-low-resource-African-LanguagesBUFFET: Benchmarking Large Language Models for Few-shot Cross-lingual Transfer

Few-Shot Cross-Lingual Transfer for Prompting Large Language Models in Low-Resource Languages

Large pre-trained language models (PLMs) are at the forefront of advances in Natural Language Proces

Zero-Shot Cross-Lingual Reranking with Large Language Models for Low-Resource Languages

Large language models (LLMs) have shown impressive zero-shot capabilities in various document rerank

Cross-lingual NER Transfer with Pretrained Language Models: Accuracy Degradation in Low-Resource Languages

Multilingual Language Models (MLLMs) exhibit robust cross-lingual transfer capabilities, or the abil

Cross-lingual transfer of multilingual models on low resource African Languages

Large multilingual models have significantly advanced natural language processing (NLP) research. Ho

karanwxliaa/Cross-lingual-transfer-of-multilingual-models-on-low-resource-African-Languages

A comparison of Cross-lingual transfer of Transformer & Neural based: Multilingual & Monolingual mod

BUFFET: Benchmarking Large Language Models for Few-shot Cross-lingual Transfer

Despite remarkable advancements in few-shot generalization in natural language processing, most mode