<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wei-2026-goblueinformatics">
    <titleInfo>
        <title>GoBlueInformatics at #SMM4H-HeaRD 2026: Long-Context Encoders and Generative Biomedical LLMs for Pathological TNM Stage Prediction</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Shangqing</namePart>
        <namePart type="family">Wei</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>We describe our systems for #SMM4H-HeaRD 2026 Task 6, which requires predicting the T, N, and M components of pathological TNM stage from TCGA pathology reports. We explored both discriminative long-context encoders and generative biomedical LLMs. For the first test set, our BioClinical-ModernBERT-large ensemble achieved 0.993 micro-F1 and 0.915 macro-F1, improving over the BB-TEN baseline scoring-log result of 0.947 micro-F1 and 0.780 macro-F1. For the harder second test set, our OpenBioLLM-8B LoRA extractor improved component macro-F1 over the organizer baseline from 0.454 to 0.626 for T, from 0.591 to 0.758 for N, and from 0.554 to 1.000 for M. These results suggest that long-context encoders are strong for explicit T and N evidence, while constrained generative LLM extraction can be effective for harder reports. The main remaining weakness is rare-class T4 recognition.</abstract>
    <identifier type="citekey">wei-2026-goblueinformatics</identifier>
    <location>
        <url>https://aclanthology.org/2026.smm4h-1.18/</url>
    </location>
    <part>
        <date>2026-07</date>
        <extent unit="page">
            <start>108</start>
            <end>112</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T GoBlueInformatics at #SMM4H-HeaRD 2026: Long-Context Encoders and Generative Biomedical LLMs for Pathological TNM Stage Prediction
%A Wei, Shangqing
%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 wei-2026-goblueinformatics
%X We describe our systems for #SMM4H-HeaRD 2026 Task 6, which requires predicting the T, N, and M components of pathological TNM stage from TCGA pathology reports. We explored both discriminative long-context encoders and generative biomedical LLMs. For the first test set, our BioClinical-ModernBERT-large ensemble achieved 0.993 micro-F1 and 0.915 macro-F1, improving over the BB-TEN baseline scoring-log result of 0.947 micro-F1 and 0.780 macro-F1. For the harder second test set, our OpenBioLLM-8B LoRA extractor improved component macro-F1 over the organizer baseline from 0.454 to 0.626 for T, from 0.591 to 0.758 for N, and from 0.554 to 1.000 for M. These results suggest that long-context encoders are strong for explicit T and N evidence, while constrained generative LLM extraction can be effective for harder reports. The main remaining weakness is rare-class T4 recognition.
%U https://aclanthology.org/2026.smm4h-1.18/
%P 108-112

Markdown (Informal)

[GoBlueInformatics at #SMM4H-HeaRD 2026: Long-Context Encoders and Generative Biomedical LLMs for Pathological TNM Stage Prediction](https://aclanthology.org/2026.smm4h-1.18/) (Wei, SMM4H 2026)

• GoBlueInformatics at #SMM4H-HeaRD 2026: Long-Context Encoders and Generative Biomedical LLMs for Pathological TNM Stage Prediction (Wei, SMM4H 2026)

ACL

• Shangqing Wei. 2026. GoBlueInformatics at #SMM4H-HeaRD 2026: Long-Context Encoders and Generative Biomedical LLMs for Pathological TNM Stage Prediction. In Proceedings of the 11th Social Media Mining for Health Research and Applications (SMM4H-HeaRD 2026) Workshop and Shared Tasks, pages 108–112, San Diego, United States. Association for Computational Linguistics.