Towards Neural No-Resource Language Translation: A Comparative Evaluation of Approaches

2024-12-29

AI Summary

This work addresses the translation challenge for “resource-poor languages”—such as Owens Valley Paiute—that lack digitized textual corpora. We propose the first research paradigm for resource-poor translation centered on zero-shot transfer, diverging from conventional low-resource machine translation approaches that rely on parallel data. Instead, we systematically investigate the feasibility of large language models (LLMs) under zero-shot conditions, comparatively evaluating three strategies: fine-tuning domain-specific models, chain-of-thought prompting, and direct prompting. We further design a language-agnostic, expert-free in-context learning (ICL) prompting scheme. Experiments demonstrate that with fewer than 100 training sentences, our method achieves BLEU scores of 0.45–0.6—matching human translation quality and substantially outperforming existing low-resource baselines. This is the first empirical validation of LLMs’ general effectiveness for translating endangered languages.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Large language models for search

Abstract

No-resource languages - those with minimal or no digital representation - pose unique challenges for machine translation (MT). Unlike low-resource languages, which rely on limited but existent corpora, no-resource languages often have fewer than 100 sentences available for training. This work explores the problem of no-resource translation through three distinct workflows: fine-tuning of translation-specific models, in-context learning with large language models (LLMs) using chain-of-reasoning prompting, and direct prompting without reasoning. Using Owens Valley Paiute as a case study, we demonstrate that no-resource translation demands fundamentally different approaches from low-resource scenarios, as traditional approaches to machine translation, such as those that work for low-resource languages, fail. Empirical results reveal that, although traditional approaches fail, the in-context learning capabilities of general-purpose large language models enable no-resource language translation that outperforms low-resource translation approaches and rivals human translations (BLEU 0.45-0.6); specifically, chain-of-reasoning prompting outperforms other methods for larger corpora, while direct prompting exhibits advantages in smaller datasets. As these approaches are language-agnostic, they have potential to be generalized to translation tasks from a wide variety of no-resource languages without expert input. These findings establish no-resource translation as a distinct paradigm requiring innovative solutions, providing practical and theoretical insights for language preservation.

Problem

• Low-resource Languages
• Machine Translation
• Internet Scarcity

Innovation

• Contextual Learning
• Ultra-low Resource Languages
• Translation Quality Enhancement