Work

Prompt Engineering Enhances Faroese MT, but Only Humans Can Tell

Year: 2025

Type: article

Abstract: This study evaluates GPT-4's English-to-Faroese translation capabilities, comparing it with multilingual models on FLORES-200 and Sprotin datasets. We propose a prompt optimization strategy using Semantic Textual Similarity (STS) to improve translation quality. Human evaluation confirms the effectiveness of STS-based few-shot example selection, though automated metrics fail to capture these improvements. Our findings advance LLM applications for low-resource language translation while highlighting the need for better evaluation methods in this context.

Source: DSpace repository (University of Tartu)

Authors: Scalvini, Barbara, Simonsen, Annika, Debess, Iben Nyholm, Einarsson, Hafsteinn

Language: English

Topic: Natural Language Processing Techniques

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W7014570581

FWCI:

Cites:

Cited by:

Related to: