<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="maraia-etal-2026-abstract">
    <titleInfo>
        <title>Abstract Activation Spaces for Content-Invariant Reasoning in Large Language Models</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Gabriele</namePart>
        <namePart type="family">Maraia</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Marco</namePart>
        <namePart type="family">Valentino</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Fabio</namePart>
        <namePart type="given">Massimo</namePart>
        <namePart type="family">Zanzotto</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Leonardo</namePart>
        <namePart type="family">Ranaldi</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>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</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">979-8-89176-459-0</identifier>
    </relatedItem>
    <abstract>LLMs often struggle with deductive judgement in syllogistic reasoning, systematically conflating semantic plausibility with formal validity—a phenomenon known as content effect. This bias persists even when models generate step-wise explanations, indicating that intermediate rationales may inherit the same semantic shortcuts that affect answers. Recent approaches mitigate this issue by increasing inference-time structural constraints, either by encouraging abstract intermediate representations or by intervening in the model’s internal computations; yet, reliably suppressing semantic interference remains an open challenge.\par To make formal deduction less sensitive to semantic content, we introduce a framework for abstraction-guided reasoning that separates structural inference from lexical semantics. We construct paired content-laden and abstract syllogisms and use the model’s activations on abstract inputs to define an abstract reasoning space. We then learn lightweight Abstractors that, from content-conditioned residual-stream states, predict representations aligned with this space and integrate these predictions via multi-layer interventions during the forward pass. Using cross-lingual transfer as a test bed, we show that abstraction-aligned steering reduces content-driven errors and improves validity-sensitive performance. Our results position activation-level abstraction as a scalable mechanism for enhancing the robustness of formal reasoning in LLMs against semantic interference.</abstract>
    <identifier type="citekey">maraia-etal-2026-abstract</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.2168/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>49295</start>
            <end>49323</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Abstract Activation Spaces for Content-Invariant Reasoning in Large Language Models
%A Maraia, Gabriele
%A Valentino, Marco
%A Zanzotto, Fabio Massimo
%A Ranaldi, Leonardo
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F maraia-etal-2026-abstract
%X LLMs often struggle with deductive judgement in syllogistic reasoning, systematically conflating semantic plausibility with formal validity—a phenomenon known as content effect. This bias persists even when models generate step-wise explanations, indicating that intermediate rationales may inherit the same semantic shortcuts that affect answers. Recent approaches mitigate this issue by increasing inference-time structural constraints, either by encouraging abstract intermediate representations or by intervening in the model’s internal computations; yet, reliably suppressing semantic interference remains an open challenge.\par To make formal deduction less sensitive to semantic content, we introduce a framework for abstraction-guided reasoning that separates structural inference from lexical semantics. We construct paired content-laden and abstract syllogisms and use the model’s activations on abstract inputs to define an abstract reasoning space. We then learn lightweight Abstractors that, from content-conditioned residual-stream states, predict representations aligned with this space and integrate these predictions via multi-layer interventions during the forward pass. Using cross-lingual transfer as a test bed, we show that abstraction-aligned steering reduces content-driven errors and improves validity-sensitive performance. Our results position activation-level abstraction as a scalable mechanism for enhancing the robustness of formal reasoning in LLMs against semantic interference.
%U https://aclanthology.org/2026.emnlp-main.2168/
%P 49295-49323

Markdown (Informal)

[Abstract Activation Spaces for Content-Invariant Reasoning in Large Language Models](https://aclanthology.org/2026.emnlp-main.2168/) (Maraia et al., EMNLP 2026)

• Abstract Activation Spaces for Content-Invariant Reasoning in Large Language Models (Maraia et al., EMNLP 2026)

ACL

• Gabriele Maraia, Marco Valentino, Fabio Massimo Zanzotto, and Leonardo Ranaldi. 2026. Abstract Activation Spaces for Content-Invariant Reasoning in Large Language Models. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 49295–49323, Budapest, Hungary. Association for Computational Linguistics.