Hire a Linguist!: Learning Endangered Languages with In-Context Linguistic Descriptions

TL;DR

This paper introduces LINGOLLM, a prompt-based method that leverages linguistic resources like dictionaries and grammar books to enable large language models to process and translate endangered languages without additional training.

Contribution

The paper presents a training-free approach that incorporates linguistic knowledge into LLM prompts, significantly improving translation performance for unseen endangered languages.

Findings

LINGOLLM improves GPT-4's BLEU score from 0 to 10.5 on endangered languages.

Using linguistic resources in prompts enhances LLM translation capabilities.

The approach demonstrates the value of linguistic knowledge for low-resource language processing.

Abstract

How can large language models (LLMs) process and translate endangered languages? Many languages lack a large corpus to train a decent LLM; therefore existing LLMs rarely perform well in unseen, endangered languages. On the contrary, we observe that 2000 endangered languages, though without a large corpus, have a grammar book or a dictionary. We propose LINGOLLM, a training-free approach to enable an LLM to process unseen languages that hardly occur in its pre-training. Our key insight is to demonstrate linguistic knowledge of an unseen language in an LLM's prompt, including a dictionary, a grammar book, and morphologically analyzed input text. We implement LINGOLLM on top of two models, GPT-4 and Mixtral, and evaluate their performance on 5 tasks across 8 endangered or low-resource languages. Our results show that LINGOLLM elevates translation capability from GPT-4's 0 to 10.5 BLEU for 10 language directions. Our findings demonstrate the tremendous value of linguistic knowledge in the age of LLMs for endangered languages. Our data, code, and model generations can be found at https://github.com/LLiLab/llm4endangeredlang.