<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="akter-etal-2026-counterfactual">
    <titleInfo>
        <title>Counterfactual Sensitivity Regularization for Trace-Sensitive Reasoning in Large Language Models</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Sanjeda</namePart>
        <namePart type="family">Akter</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ibne</namePart>
        <namePart type="given">Farabi</namePart>
        <namePart type="family">Shihab</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Anuj</namePart>
        <namePart type="family">Sharma</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-10</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics: EMNLP 2026</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Hung-Yi</namePart>
            <namePart type="family">Lee</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Sunipa</namePart>
            <namePart type="family">Dev</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jessy</namePart>
            <namePart type="family">Li</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Desmond</namePart>
            <namePart type="family">Elliott</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Budapest, Hungary</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-462-0</identifier>
    </relatedItem>
    <abstract>Large language models can produce correct final answers with reasoning traces that are incomplete, inconsistent, or weakly coupled to the answer. We study whether training can increase this coupling in structured reasoning domains. We introduce Counterfactual Sensitivity Regularization (CSR), a training objective that creates verifier-approved counterfactual edits to reasoning traces, such as changing an arithmetic operator or a logical relation, and penalizes models when the answer distribution remains unchanged after an edit that should change the computation. We evaluate this behavior with Counterfactual Outcome Sensitivity (COS), the fraction of originally correct examples on which a validity-breaking trace edit changes the final answer. Because COS is aligned with the CSR objective by construction, we report it jointly with counterfactual target accuracy, target coverage, and null-edit preservation, and alongside evidence collected outside the training loop: natural-unfaithfulness audits, external formal verification, input contrast sets, and trace-dependence probes. Across arithmetic, formal logic, multi-hop question answering, and biomedical QA, CSR substantially improves edit sensitivity while incurring small accuracy trade-offs. The gains are largest in domains where the edited operators can be identified and verified reliably; performance degrades as verifier precision falls. These results support CSR as a practical regularizer for trace-sensitive reasoning, not as a complete solution to explanation faithfulness.</abstract>
    <identifier type="citekey">akter-etal-2026-counterfactual</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-emnlp.539/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>11320</start>
            <end>11352</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Counterfactual Sensitivity Regularization for Trace-Sensitive Reasoning in Large Language Models
%A Akter, Sanjeda
%A Shihab, Ibne Farabi
%A Sharma, Anuj
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F akter-etal-2026-counterfactual
%X Large language models can produce correct final answers with reasoning traces that are incomplete, inconsistent, or weakly coupled to the answer. We study whether training can increase this coupling in structured reasoning domains. We introduce Counterfactual Sensitivity Regularization (CSR), a training objective that creates verifier-approved counterfactual edits to reasoning traces, such as changing an arithmetic operator or a logical relation, and penalizes models when the answer distribution remains unchanged after an edit that should change the computation. We evaluate this behavior with Counterfactual Outcome Sensitivity (COS), the fraction of originally correct examples on which a validity-breaking trace edit changes the final answer. Because COS is aligned with the CSR objective by construction, we report it jointly with counterfactual target accuracy, target coverage, and null-edit preservation, and alongside evidence collected outside the training loop: natural-unfaithfulness audits, external formal verification, input contrast sets, and trace-dependence probes. Across arithmetic, formal logic, multi-hop question answering, and biomedical QA, CSR substantially improves edit sensitivity while incurring small accuracy trade-offs. The gains are largest in domains where the edited operators can be identified and verified reliably; performance degrades as verifier precision falls. These results support CSR as a practical regularizer for trace-sensitive reasoning, not as a complete solution to explanation faithfulness.
%U https://aclanthology.org/2026.findings-emnlp.539/
%P 11320-11352

Markdown (Informal)

[Counterfactual Sensitivity Regularization for Trace-Sensitive Reasoning in Large Language Models](https://aclanthology.org/2026.findings-emnlp.539/) (Akter et al., Findings 2026)

• Counterfactual Sensitivity Regularization for Trace-Sensitive Reasoning in Large Language Models (Akter et al., Findings 2026)

ACL

• Sanjeda Akter, Ibne Farabi Shihab, and Anuj Sharma. 2026. Counterfactual Sensitivity Regularization for Trace-Sensitive Reasoning in Large Language Models. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 11320–11352, Budapest, Hungary. Association for Computational Linguistics.