Year: 2024
Type: preprint
Abstract: Large language models (LLMs) have advanced the state of the art in natural language processing. However, their predominant design for English or a limited set of languages creates a substantial gap in their effectiveness for low-resource languages. To bridge this gap, we introduce MaLA-500, a novel large language model designed to cover an extensive range of 534 languages. To train MaLA-500, we employ vocabulary extension and continued pretraining on LLaMA 2 with Glot500-c. Our intrinsic evaluation demonstrates that MaLA-500 is better at predicting the given texts of low-resource languages than existing multilingual LLMs. Moreover, the extrinsic evaluation of in-context learning shows that MaLA-500 outperforms previous LLMs on SIB200 and Taxi1500 by a significant margin, i.e., 11.68% and 4.82% marco-average accuracy across languages. We release MaLA-500 at https://huggingface.co/MaLA-LM
Source: arXiv (Cornell University)
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Awards: Deep Learning for Structured Prediction in Natural Language Processing, Found in Translation – Natural Language Understanding with Cross-Lingual Grounding, Non-sequence models for tokenization replacement, G4459410915, High Performance Language Technologies, Unified Transcription and Translation for Extended Reality, G612072080, European Learning and Intelligent Systems Excellence, Deep Learning for Structured Prediction in Natural Language Processing
OpenAlex ID: W4391244633