@inproceedings{zeidi-etal-2026-pei,
title = "{PEI} at {\#}{SMM}4{H}-{H}ea{RD} 2026: Enhancing Patient Metadata Detection via Hypothesis-Conditioned Classification and Paraphrase-Based Data Augmentation",
author = {Zeidi, Farnaz and
Christof, Roman and
Zeidi, Farnoush and
K{\"o}nig, Renate and
Childs, Liam},
editor = "Lopez-Garcia, Guillermo and
Gonzalez-Hernandez, Graciela",
booktitle = "Proceedings of the 11th Social Media Mining for Health Research and Applications ({SMM}4{H}-{H}ea{RD} 2026) Workshop and Shared Tasks",
month = jul,
year = "2026",
address = "San Diego, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.smm4h-1.24/",
pages = "146--153",
ISBN = "979-8-89176-432-3",
abstract = "This paper presents our approach to Task 5 of the {\#}SMM4H-HeaRD 2026 Workshop, which focuses on detecting patient metadata in SARS-CoV-2 sequencing articles as a binary classification task. We explore both encoder-based and large language model (LLM) approaches, using BioM-BERT as a baseline and Mistral-Nemo as the LLM. To improve performance, we propose a paraphrase-based data augmentation pipeline using Qwen3, where paraphrased training and validation instances are added for fine-tuning. For the LLM, we perform prompt refinement and error analysis, while for the encoder-based model, we reformulate the task as a hypothesis-conditioned classification task inspired by Natural Language Inference (NLI). Our methods improve both models: Mistral-Nemo increases from 0.423 to 0.750 F1, and BioM-BERT from 0.801 to 0.821 on the validation set. Although Mistral-Nemo does not surpass BioM-BERT, our best BioM-BERT model achieves an F1-score of 0.786 on the test set, outperforming the mean and median of competing systems. To support reproducibility, we release our best-performing model on Hugging Face."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zeidi-etal-2026-pei">
<titleInfo>
<title>PEI at #SMM4H-HeaRD 2026: Enhancing Patient Metadata Detection via Hypothesis-Conditioned Classification and Paraphrase-Based Data Augmentation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Farnaz</namePart>
<namePart type="family">Zeidi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Roman</namePart>
<namePart type="family">Christof</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Farnoush</namePart>
<namePart type="family">Zeidi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Renate</namePart>
<namePart type="family">König</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Liam</namePart>
<namePart type="family">Childs</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 11th Social Media Mining for Health Research and Applications (SMM4H-HeaRD 2026) Workshop and Shared Tasks</title>
</titleInfo>
<name type="personal">
<namePart type="given">Guillermo</namePart>
<namePart type="family">Lopez-Garcia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Graciela</namePart>
<namePart type="family">Gonzalez-Hernandez</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-432-3</identifier>
</relatedItem>
<abstract>This paper presents our approach to Task 5 of the #SMM4H-HeaRD 2026 Workshop, which focuses on detecting patient metadata in SARS-CoV-2 sequencing articles as a binary classification task. We explore both encoder-based and large language model (LLM) approaches, using BioM-BERT as a baseline and Mistral-Nemo as the LLM. To improve performance, we propose a paraphrase-based data augmentation pipeline using Qwen3, where paraphrased training and validation instances are added for fine-tuning. For the LLM, we perform prompt refinement and error analysis, while for the encoder-based model, we reformulate the task as a hypothesis-conditioned classification task inspired by Natural Language Inference (NLI). Our methods improve both models: Mistral-Nemo increases from 0.423 to 0.750 F1, and BioM-BERT from 0.801 to 0.821 on the validation set. Although Mistral-Nemo does not surpass BioM-BERT, our best BioM-BERT model achieves an F1-score of 0.786 on the test set, outperforming the mean and median of competing systems. To support reproducibility, we release our best-performing model on Hugging Face.</abstract>
<identifier type="citekey">zeidi-etal-2026-pei</identifier>
<location>
<url>https://aclanthology.org/2026.smm4h-1.24/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>146</start>
<end>153</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T PEI at #SMM4H-HeaRD 2026: Enhancing Patient Metadata Detection via Hypothesis-Conditioned Classification and Paraphrase-Based Data Augmentation
%A Zeidi, Farnaz
%A Christof, Roman
%A Zeidi, Farnoush
%A König, Renate
%A Childs, Liam
%Y Lopez-Garcia, Guillermo
%Y Gonzalez-Hernandez, Graciela
%S Proceedings of the 11th Social Media Mining for Health Research and Applications (SMM4H-HeaRD 2026) Workshop and Shared Tasks
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, United States
%@ 979-8-89176-432-3
%F zeidi-etal-2026-pei
%X This paper presents our approach to Task 5 of the #SMM4H-HeaRD 2026 Workshop, which focuses on detecting patient metadata in SARS-CoV-2 sequencing articles as a binary classification task. We explore both encoder-based and large language model (LLM) approaches, using BioM-BERT as a baseline and Mistral-Nemo as the LLM. To improve performance, we propose a paraphrase-based data augmentation pipeline using Qwen3, where paraphrased training and validation instances are added for fine-tuning. For the LLM, we perform prompt refinement and error analysis, while for the encoder-based model, we reformulate the task as a hypothesis-conditioned classification task inspired by Natural Language Inference (NLI). Our methods improve both models: Mistral-Nemo increases from 0.423 to 0.750 F1, and BioM-BERT from 0.801 to 0.821 on the validation set. Although Mistral-Nemo does not surpass BioM-BERT, our best BioM-BERT model achieves an F1-score of 0.786 on the test set, outperforming the mean and median of competing systems. To support reproducibility, we release our best-performing model on Hugging Face.
%U https://aclanthology.org/2026.smm4h-1.24/
%P 146-153
Markdown (Informal)
[PEI at #SMM4H-HeaRD 2026: Enhancing Patient Metadata Detection via Hypothesis-Conditioned Classification and Paraphrase-Based Data Augmentation](https://aclanthology.org/2026.smm4h-1.24/) (Zeidi et al., SMM4H 2026)
ACL