@inproceedings{syed-pedersen-2026-duluth,
title = "{D}uluth at {S}em{E}val-2026 Task 6: {D}e{BERT}a with {LLM}-Augmented Data for Unmasking Political Question Evasions",
author = "Syed, Shujauddin and
Pedersen, Ted",
editor = "Kochmar, Ekaterina and
Ghosh, Debanjan and
North, Kai and
Komachi, Mamoru and
Zampieri, Marcos",
booktitle = "Proceedings of the 20th {I}nternational {W}orkshop on {S}emantic {E}valuation (2026)",
month = jul,
year = "2026",
address = "San Diego, California, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.semeval-1.94/",
doi = "10.18653/v1/2026.semeval-1.94",
pages = "650--657",
ISBN = "979-8-89176-414-9",
abstract = "This paper presents the Duluth approach to SemEval-2026 Task 6 on CLARITY: Unmasking Political Question Evasions. We address Task 1 (clarity-level classification) and Task 2 (evasion-level classification), both of which involve classifying question{--}answer pairs from U.S. presidential interviews using a two-level taxonomy of response clarity. Our system is based on DeBERTa-V3-base, extended with focal loss, layer-wise learning rate decay, and boolean discourse features. To address class imbalance in the training data, we augment minority classes using synthetic examples generated by Gemini 3 and Claude Sonnet 4.5. Our best configuration achieved a Macro F1 of 0.76 on the Task 1 evaluation set, placing 8th out of 40 teams. The top-ranked system (TeleAI) achieved 0.89, while the mean score across participants was 0.70. Error analysis reveals that the dominant source of misclassification is confusion between Ambivalent and Clear Reply responses, a pattern that mirrors disagreements among human annotators. Our findings demonstrate that LLM-based data augmentation can meaningfully improve minority-class recall on nuanced political discourse tasks."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="syed-pedersen-2026-duluth">
<titleInfo>
<title>Duluth at SemEval-2026 Task 6: DeBERTa with LLM-Augmented Data for Unmasking Political Question Evasions</title>
</titleInfo>
<name type="personal">
<namePart type="given">Shujauddin</namePart>
<namePart type="family">Syed</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ted</namePart>
<namePart type="family">Pedersen</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 20th International Workshop on Semantic Evaluation (2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Kochmar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Debanjan</namePart>
<namePart type="family">Ghosh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kai</namePart>
<namePart type="family">North</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mamoru</namePart>
<namePart type="family">Komachi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marcos</namePart>
<namePart type="family">Zampieri</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, USA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-414-9</identifier>
</relatedItem>
<abstract>This paper presents the Duluth approach to SemEval-2026 Task 6 on CLARITY: Unmasking Political Question Evasions. We address Task 1 (clarity-level classification) and Task 2 (evasion-level classification), both of which involve classifying question–answer pairs from U.S. presidential interviews using a two-level taxonomy of response clarity. Our system is based on DeBERTa-V3-base, extended with focal loss, layer-wise learning rate decay, and boolean discourse features. To address class imbalance in the training data, we augment minority classes using synthetic examples generated by Gemini 3 and Claude Sonnet 4.5. Our best configuration achieved a Macro F1 of 0.76 on the Task 1 evaluation set, placing 8th out of 40 teams. The top-ranked system (TeleAI) achieved 0.89, while the mean score across participants was 0.70. Error analysis reveals that the dominant source of misclassification is confusion between Ambivalent and Clear Reply responses, a pattern that mirrors disagreements among human annotators. Our findings demonstrate that LLM-based data augmentation can meaningfully improve minority-class recall on nuanced political discourse tasks.</abstract>
<identifier type="citekey">syed-pedersen-2026-duluth</identifier>
<identifier type="doi">10.18653/v1/2026.semeval-1.94</identifier>
<location>
<url>https://aclanthology.org/2026.semeval-1.94/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>650</start>
<end>657</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Duluth at SemEval-2026 Task 6: DeBERTa with LLM-Augmented Data for Unmasking Political Question Evasions
%A Syed, Shujauddin
%A Pedersen, Ted
%Y Kochmar, Ekaterina
%Y Ghosh, Debanjan
%Y North, Kai
%Y Komachi, Mamoru
%Y Zampieri, Marcos
%S Proceedings of the 20th International Workshop on Semantic Evaluation (2026)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, USA
%@ 979-8-89176-414-9
%F syed-pedersen-2026-duluth
%X This paper presents the Duluth approach to SemEval-2026 Task 6 on CLARITY: Unmasking Political Question Evasions. We address Task 1 (clarity-level classification) and Task 2 (evasion-level classification), both of which involve classifying question–answer pairs from U.S. presidential interviews using a two-level taxonomy of response clarity. Our system is based on DeBERTa-V3-base, extended with focal loss, layer-wise learning rate decay, and boolean discourse features. To address class imbalance in the training data, we augment minority classes using synthetic examples generated by Gemini 3 and Claude Sonnet 4.5. Our best configuration achieved a Macro F1 of 0.76 on the Task 1 evaluation set, placing 8th out of 40 teams. The top-ranked system (TeleAI) achieved 0.89, while the mean score across participants was 0.70. Error analysis reveals that the dominant source of misclassification is confusion between Ambivalent and Clear Reply responses, a pattern that mirrors disagreements among human annotators. Our findings demonstrate that LLM-based data augmentation can meaningfully improve minority-class recall on nuanced political discourse tasks.
%R 10.18653/v1/2026.semeval-1.94
%U https://aclanthology.org/2026.semeval-1.94/
%U https://doi.org/10.18653/v1/2026.semeval-1.94
%P 650-657
Markdown (Informal)
[Duluth at SemEval-2026 Task 6: DeBERTa with LLM-Augmented Data for Unmasking Political Question Evasions](https://aclanthology.org/2026.semeval-1.94/) (Syed & Pedersen, SemEval 2026)
ACL