Can Uniform Meaning Representation Help GPT-4 Translate from Indigenous Languages?

TL;DR

This paper investigates whether incorporating Uniform Meaning Representation (UMR) into GPT-4 prompts can improve translation performance for indigenous languages, showing promising results in low-resource language translation tasks.

Contribution

It demonstrates that integrating UMR into GPT-4 prompts can significantly enhance translation accuracy for indigenous languages, a novel application of semantic representations in large language models.

Findings

UMR integration improves translation performance in most test cases.

Statistically significant performance gains observed with UMR prompts.

Potential for UMR to aid low-resource language processing in LLMs.

Abstract

While ChatGPT and GPT-based models are able to effectively perform many tasks without additional fine-tuning, they struggle with tasks related to extremely low-resource languages and indigenous languages. Uniform Meaning Representation (UMR), a semantic representation designed to capture the meaning of texts in many languages, is well-positioned to be leveraged in the development of low-resource language technologies. In this work, we explore the downstream utility of UMR for low-resource languages by incorporating it into GPT-4 prompts. Specifically, we examine the ability of GPT-4 to perform translation from three indigenous languages (Navajo, Ar\'apaho, and Kukama), with and without demonstrations, as well as with and without UMR annotations. Ultimately, we find that in the majority of our test cases, integrating UMR into the prompt results in a statistically significant increase in performance, which is a promising indication of future applications of the UMR formalism.