Named Entity Recognition in the Era of Large Language Models: A Comparative Study

Named Entity Recognition (NER) is a fundamental task in Natural Language Processing (NLP), focused on identifying and classifying entities such as persons, locations, and organizations in unstructured text. The advent of Large Language Models (LLMs) such as GPT-4 has brought significant advancements to NER, yet their performance compared to traditional or task-specific models remains underexplored. This paper presents a comprehensive comparative analysis of traditional approaches, such as CRFs and LSTMs, and modern frameworks, including LLMs. Using ten diverse datasets spanning general, biomedical, and domain-specific contexts, this study evaluates the strengths, limitations, and adaptability of these models.