Evaluating Gender Bias on Machine Translation using Large Language Models

This thesis evaluates gender bias in Machine Translation using Large Language Models. A comprehensive benchmarking process assessed various encoder-decoder NMT models and decoder-only base LLMs for English-to-Catalan and English-to-Spanish translations. Distinct test sets, including FLoRes, WinoMT,...

Descripción completa

Detalles Bibliográficos
Autor: Sant Savall, Aleix
Tipo de recurso: tesis de maestría
Fecha de publicación:2024
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/410885
Acceso en línea:https://hdl.handle.net/2117/410885
Access Level:acceso abierto
Palabra clave:Machine translating
Large Language Models
Machine Translation
Gender Bias
Prompting Engineering
Traducció automàtica
Àrees temàtiques de la UPC::Enginyeria de la telecomunicació::Processament del senyal::Processament de la parla i del senyal acústic
Descripción
Sumario:This thesis evaluates gender bias in Machine Translation using Large Language Models. A comprehensive benchmarking process assessed various encoder-decoder NMT models and decoder-only base LLMs for English-to-Catalan and English-to-Spanish translations. Distinct test sets, including FLoRes, WinoMT, Gold BUG, and MuST-SHE, were employed to evaluate translation quality and gender bias, obtaining metrics such as BLEU, COMET, Gender Accuracy, F1-male, F1-female, ?G, and ?S. Results indicated the presence of gender bias across all models, with base LLMs exhibiting more bias than NMT models. To mitigate this bias, prompting engineering techniques were implemented in an instruction-tuned LLM. After multiple attempts, a prompt resulting in a significant reduction in gender bias across the test sets was identified, achieving remarkable gender scores. The prompt followed a simplified step-by-step approach with 5-shots on anti-stereotypical content and increased female representation. This prompt enabled the instructed LLM to perform competitively, achieving results comparable to NMT models, even surpassing some of them. However, a loss in translation quality using this prompt was observed.