<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="roy-malhotra-2026-modusponens">
    <titleInfo>
        <title>ModusPonens at SemEval-2026 Task 11: Breaking the Plausibility Trap in LLMs via Conflict-Aware Ensembling</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Soumyajit</namePart>
        <namePart type="family">Roy</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Manav</namePart>
        <namePart type="family">Malhotra</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>
        <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>Large Language Models (LLMs) often struggle to disentangle formal logical validity from real-world plausibility, a phenomenon known as the “belief bias”. This paper describes our submission to SemEval-2026 Task 11. We frame the task as a calibration problem between “System 1” (heuristic) and “System 2” (logical) thinking. Our experiments reveal that standard neuro-symbolic interventions, such as Structural Chain-of-Thought (CoT) and Nonsense Augmentation, degrade performance in low-resource regimes due to an “abstraction penalty”. Instead, we propose a Conflict-Aware Logit Ensemble. We fine-tune two variations of Qwen-2.5-14B: a standard “Believer” model and a bias-hardened “Skeptic” model trained on oversampled conflict data. By ensembling their logits via soft-voting, we achieve a Pareto-optimal balance, reducing the Total Content Effect (TCE) to 3.21 while maintaining an overall accuracy of 94.27%, resulting in a Combined Score of 39.09.</abstract>
    <identifier type="citekey">roy-malhotra-2026-modusponens</identifier>
    <location>
        <url>https://aclanthology.org/2026.semeval-1.10/</url>
    </location>
    <part>
        <date>2026-07</date>
        <extent unit="page">
            <start>65</start>
            <end>71</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T ModusPonens at SemEval-2026 Task 11: Breaking the Plausibility Trap in LLMs via Conflict-Aware Ensembling
%A Roy, Soumyajit
%A Malhotra, Manav
%Y Kochmar, Ekaterina
%Y Ghosh, Debanjan
%Y North, Kai
%Y Komachi, Mamoru
%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 roy-malhotra-2026-modusponens
%X Large Language Models (LLMs) often struggle to disentangle formal logical validity from real-world plausibility, a phenomenon known as the “belief bias”. This paper describes our submission to SemEval-2026 Task 11. We frame the task as a calibration problem between “System 1” (heuristic) and “System 2” (logical) thinking. Our experiments reveal that standard neuro-symbolic interventions, such as Structural Chain-of-Thought (CoT) and Nonsense Augmentation, degrade performance in low-resource regimes due to an “abstraction penalty”. Instead, we propose a Conflict-Aware Logit Ensemble. We fine-tune two variations of Qwen-2.5-14B: a standard “Believer” model and a bias-hardened “Skeptic” model trained on oversampled conflict data. By ensembling their logits via soft-voting, we achieve a Pareto-optimal balance, reducing the Total Content Effect (TCE) to 3.21 while maintaining an overall accuracy of 94.27%, resulting in a Combined Score of 39.09.
%U https://aclanthology.org/2026.semeval-1.10/
%P 65-71

Markdown (Informal)

[ModusPonens at SemEval-2026 Task 11: Breaking the Plausibility Trap in LLMs via Conflict-Aware Ensembling](https://aclanthology.org/2026.semeval-1.10/) (Roy & Malhotra, SemEval 2026)

• ModusPonens at SemEval-2026 Task 11: Breaking the Plausibility Trap in LLMs via Conflict-Aware Ensembling (Roy & Malhotra, SemEval 2026)

ACL

• Soumyajit Roy and Manav Malhotra. 2026. ModusPonens at SemEval-2026 Task 11: Breaking the Plausibility Trap in LLMs via Conflict-Aware Ensembling. In Proceedings of the 20th International Workshop on Semantic Evaluation (2026), pages 65–71, San Diego, California, USA. Association for Computational Linguistics.