<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="van-der-weijden-etal-2026-reproducing">
    <titleInfo>
        <title>Reproducing and Evaluating the Generalizability of Subliminal Learning in Open-Weight Models</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Daan</namePart>
        <namePart type="family">van der Weijden</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Nathan</namePart>
        <namePart type="family">Brack</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Selene</namePart>
        <namePart type="family">Baez Santamaria</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 9th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Dana</namePart>
            <namePart type="family">Arad</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Yonatan</namePart>
            <namePart type="family">Belinkov</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Hanjie</namePart>
            <namePart type="family">Chen</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Nils</namePart>
            <namePart type="family">Feldhus</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Aaron</namePart>
            <namePart type="family">Mueller</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Hosein</namePart>
            <namePart type="family">Mohebbi</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Gabriele</namePart>
            <namePart type="family">Sarti</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Martin</namePart>
            <namePart type="family">Tutek</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">979-8-89176-476-7</identifier>
    </relatedItem>
    <abstract>In this reproduction paper we investigate subliminal learning, a consequence of distillation where teacher models transmit behavioral preference traits through semantically unrelated data. The original paper explores two types of traits (animal preferences and misalignment), three data modalities (number sequences, code, and chain of thought), and several model families. We reproduce their experiments and extend the setup along three axes: new preference categories (actors and politicians), a new task (chess move generation), and an additional open-weight model (Ministral8B). We also run a controlled ablation on the numbers task’s answer-space size (1-, 2-, and 3-digit sequences). We focus on open-weight models with accessible checkpoints on HuggingFace, since the original paper’s GPT-4.x fine-tuning is no longer available. Our reproduction supports the original paper’s claims, but our extensions show they are not universal as transmission strength varies across traits and tasks, and one model shows almost no effect at all.</abstract>
    <identifier type="citekey">van-der-weijden-etal-2026-reproducing</identifier>
    <location>
        <url>https://aclanthology.org/2026.blackboxnlp-1.62/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>1113</start>
            <end>1120</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Reproducing and Evaluating the Generalizability of Subliminal Learning in Open-Weight Models
%A van der Weijden, Daan
%A Brack, Nathan
%A Baez Santamaria, Selene
%Y Arad, Dana
%Y Belinkov, Yonatan
%Y Chen, Hanjie
%Y Feldhus, Nils
%Y Mueller, Aaron
%Y Mohebbi, Hosein
%Y Sarti, Gabriele
%Y Tutek, Martin
%S Proceedings of the 9th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-476-7
%F van-der-weijden-etal-2026-reproducing
%X In this reproduction paper we investigate subliminal learning, a consequence of distillation where teacher models transmit behavioral preference traits through semantically unrelated data. The original paper explores two types of traits (animal preferences and misalignment), three data modalities (number sequences, code, and chain of thought), and several model families. We reproduce their experiments and extend the setup along three axes: new preference categories (actors and politicians), a new task (chess move generation), and an additional open-weight model (Ministral8B). We also run a controlled ablation on the numbers task’s answer-space size (1-, 2-, and 3-digit sequences). We focus on open-weight models with accessible checkpoints on HuggingFace, since the original paper’s GPT-4.x fine-tuning is no longer available. Our reproduction supports the original paper’s claims, but our extensions show they are not universal as transmission strength varies across traits and tasks, and one model shows almost no effect at all.
%U https://aclanthology.org/2026.blackboxnlp-1.62/
%P 1113-1120

Markdown (Informal)

[Reproducing and Evaluating the Generalizability of Subliminal Learning in Open-Weight Models](https://aclanthology.org/2026.blackboxnlp-1.62/) (van der Weijden et al., BlackboxNLP 2026)

• Reproducing and Evaluating the Generalizability of Subliminal Learning in Open-Weight Models (van der Weijden et al., BlackboxNLP 2026)

ACL

• Daan van der Weijden, Nathan Brack, and Selene Baez Santamaria. 2026. Reproducing and Evaluating the Generalizability of Subliminal Learning in Open-Weight Models. In Proceedings of the 9th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP, pages 1113–1120, Budapest, Hungary. Association for Computational Linguistics.