Dissertação de Mestrado

Exploring LLM-based Approaches for Open-ended Named Entity Recognition in Low-resource Domains

Miguel Angelo Medeiros Rebocho — 2025

Informações chave

Autores:

Miguel Angelo Medeiros Rebocho (Miguel Angelo Medeiros Rebocho)

Orientadores:

Bruno Emanuel Da Graça Martins (Bruno Emanuel Da Graça Martins); Manuel Filipe Silva Goulão

Publicado em

12 de dezembro de 2025

Resumo

This work explores how Large Language Models (LLMs) can perform Named Entity Recognition (NER) in low-resource and domain-specific scenarios, focusing on the legal domain and Portuguese language. We present a modular framework that separates and evaluates the effect of entity information, in-context demonstrations, reasoning instructions, and error reflection, across 10 datasets. We conduct systematic experiments with all 10 datasets, demonstrating that smaller, cost-efficient models like GPT-4o mini can achieve competitive results when paired with optimized prompting and retrieval. Our final configuration, using representative central example spans and 10% of the training database as in-context demonstrations, retrieved via cosine similarity, achieves an average micro-F1 of 73.5%, showcasing that our refined prompting approaches and retrieval can offset the need for fine-tuning in resource-constrained NER settings.

Detalhes da publicação

Autores da comunidade :

Orientadores desta instituição:

Domínio Científico (FOS)

electrical-engineering-electronic-engineering-information-engineering - Engenharia Eletrotécnica, Eletrónica e Informática

Idioma da publicação (código ISO)

eng - Inglês

Acesso à publicação:

Embargo levantado

Data do fim do embargo:

12 de outubro de 2026

Nome da instituição

Instituto Superior Técnico