# FMI@SU ToxHabits: Evaluating LLMs Performance on Toxic Habit Extraction in Spanish Clinical Texts

arXiv: https://arxiv.org/abs/2604.06403 · HTML: https://mlsanity.com/paper/2604.06403/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 16 references that are in the map.
- First posted on arXiv 7 Apr 2026.

## Builds on (most cited first, 7 of 7)

- [UniversalNER: Targeted Distillation from Large Language Models for Open Named Entity Recognition](https://mlsanity.com/paper/2308.03279/) (2023)
- [Exploring the Effectiveness of Instruction Tuning in Biomedical Language Processing](https://mlsanity.com/paper/2401.00579/) (2023)
- [LLMs in Biomedicine: A study on clinical Named Entity Recognition](https://mlsanity.com/paper/2404.07376/) (2024)
- [Inspire the Large Language Model by External Knowledge on BioMedical Named Entity Recognition](https://mlsanity.com/paper/2309.12278/) (2023)
- [Large Language Models Struggle in Token-Level Clinical Named Entity Recognition](https://mlsanity.com/paper/2407.00731/) (2024)
- [A Comparative Study of DSPy Teleprompter Algorithms for Aligning Large Language Models Evaluation Metrics to Human Evaluation](https://mlsanity.com/paper/2412.15298/) (2024)
- [VANER: Leveraging Large Language Model for Versatile and Adaptive Biomedical Named Entity Recognition](https://mlsanity.com/paper/2404.17835/) (2024)

## Built on by (most cited first, 1 of 1)

- [Object Aligner: A Configurable JSON Schema Similarity Score for Graphs, Applied to LLM Prompt Optimization](https://mlsanity.com/paper/2607.01972/) (2026)

## Abstract (arXiv)

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.