<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="huiskens-etal-2026-dolle">
    <titleInfo>
        <title>d’Olle Grieze at SemEval-2026 Task 11: Comparing the Impact of Supervised Fine-Tuning and Activation Steering on Mitigating Content Effect Bias in Syllogistic Reasoning</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Twan</namePart>
        <namePart type="family">Huiskens</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Tian</namePart>
        <namePart type="family">Niezing</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Koen</namePart>
        <namePart type="family">Snelten</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>We investigate the content effect bias in Large Language Models (LLMs) as part of SemEval 2026 Task 11. We compare the impact of supervised fine-tuning using low-rank adaptation against activation steering across several model families, including LLaMA, Gemma and Qwen. Our results show that SFT improves accuracy, with LLaMa 8B reaching 98.75\textbackslash% accuracy. Activation steering offers limited effectiveness in mitigating the content effect bias. A logit lens analysis further reveals that fine-tuning successfully shifts the model’s focus toward logical structure, specifically within the later layers.</abstract>
    <identifier type="citekey">huiskens-etal-2026-dolle</identifier>
    <location>
        <url>https://aclanthology.org/2026.semeval-1.167/</url>
    </location>
    <part>
        <date>2026-07</date>
        <extent unit="page">
            <start>1268</start>
            <end>1281</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T d’Olle Grieze at SemEval-2026 Task 11: Comparing the Impact of Supervised Fine-Tuning and Activation Steering on Mitigating Content Effect Bias in Syllogistic Reasoning
%A Huiskens, Twan
%A Niezing, Tian
%A Snelten, Koen
%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 huiskens-etal-2026-dolle
%X We investigate the content effect bias in Large Language Models (LLMs) as part of SemEval 2026 Task 11. We compare the impact of supervised fine-tuning using low-rank adaptation against activation steering across several model families, including LLaMA, Gemma and Qwen. Our results show that SFT improves accuracy, with LLaMa 8B reaching 98.75\textbackslash% accuracy. Activation steering offers limited effectiveness in mitigating the content effect bias. A logit lens analysis further reveals that fine-tuning successfully shifts the model’s focus toward logical structure, specifically within the later layers.
%U https://aclanthology.org/2026.semeval-1.167/
%P 1268-1281

Markdown (Informal)

[d’Olle Grieze at SemEval-2026 Task 11: Comparing the Impact of Supervised Fine-Tuning and Activation Steering on Mitigating Content Effect Bias in Syllogistic Reasoning](https://aclanthology.org/2026.semeval-1.167/) (Huiskens et al., SemEval 2026)

• d’Olle Grieze at SemEval-2026 Task 11: Comparing the Impact of Supervised Fine-Tuning and Activation Steering on Mitigating Content Effect Bias in Syllogistic Reasoning (Huiskens et al., SemEval 2026)

ACL

• Twan Huiskens, Tian Niezing, and Koen Snelten. 2026. d’Olle Grieze at SemEval-2026 Task 11: Comparing the Impact of Supervised Fine-Tuning and Activation Steering on Mitigating Content Effect Bias in Syllogistic Reasoning. In Proceedings of the 20th International Workshop on Semantic Evaluation (2026), pages 1268–1281, San Diego, California, USA. Association for Computational Linguistics.