FMI@SU ToxHabits: Evaluating LLMs Performance on Toxic Habit Extraction in Spanish Clinical Texts
This work addresses the challenge of named entity recognition for toxic habits in non-English clinical data, representing an incremental improvement in domain-specific applications.
The paper tackled the problem of extracting toxic habit mentions from Spanish clinical texts, achieving an F1 score of 0.65 using GPT-4.1 with few-shot prompting.
The paper presents an approach for the recognition of toxic habits named entities in Spanish clinical texts. The approach was developed for the ToxHabits Shared Task. Our team participated in subtask 1, which aims to detect substance use and abuse mentions in clinical case reports and classify them in four categories (Tobacco, Alcohol, Cannabis, and Drug). We explored various methods of utilizing LLMs for the task, including zero-shot, few-shot, and prompt optimization, and found that GPT-4.1's few-shot prompting performed the best in our experiments. Our method achieved an F1 score of 0.65 on the test set, demonstrating a promising result for recognizing named entities in languages other than English.