FMI@SU ToxHabits: Evaluating LLMs Performance on Toxic Habit Extraction in Spanish Clinical Texts
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866918433270005760 |
|---|---|
| author | Vassileva, Sylvia Koychev, Ivan Boytcheva, Svetla |
| author_facet | Vassileva, Sylvia Koychev, Ivan Boytcheva, Svetla |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_06403 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | FMI@SU ToxHabits: Evaluating LLMs Performance on Toxic Habit Extraction in Spanish Clinical Texts Vassileva, Sylvia Koychev, Ivan Boytcheva, Svetla Computation and Language Artificial Intelligence 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. |
| title | FMI@SU ToxHabits: Evaluating LLMs Performance on Toxic Habit Extraction in Spanish Clinical Texts |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2604.06403 |