<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="brasoveanu-etal-2026-relational">
    <titleInfo>
        <title>Relational Attention for Data-Efficient Language Modeling</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Adrian</namePart>
        <namePart type="family">Brasoveanu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ece</namePart>
        <namePart type="family">Takmaz</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jakub</namePart>
        <namePart type="family">Dotlacil</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 Second BabyLM Workshop</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Leshem</namePart>
            <namePart type="family">Choshen</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Ryan</namePart>
            <namePart type="family">Cotterell</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Mustafa</namePart>
            <namePart type="given">Omer</namePart>
            <namePart type="family">Gul</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jaap</namePart>
            <namePart type="family">Jumelet</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Tal</namePart>
            <namePart type="family">Linzen</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">Suchir</namePart>
            <namePart type="family">Salhan</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Raj</namePart>
            <namePart type="given">Sanjay</namePart>
            <namePart type="family">Shah</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Alex</namePart>
            <namePart type="family">Warstadt</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Ethan</namePart>
            <namePart type="given">Gotlieb</namePart>
            <namePart type="family">Wilcox</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">979-8-89176-475-0</identifier>
    </relatedItem>
    <abstract>We present Relational BabyLM, a system submission to the BabyLM 2026 challenge that combines two cognitively motivated inductive biases in a single decoder-only Trans- former. Architecturally, we replace standard self-attention with a Dual Attention Trans- former (DAT), which separates the routing of object-level (“sensory”) lexical features from structural/relational information (Altabaa and Lafferty, 2025; Altabaa et al., 2024; Webb et al., 2024; Kerg et al., 2022; Webb et al., 2021). Relational attention (RA) disentangled from self-attention greatly increases data efficiency and out-of-training-sample generalization on purely relational tasks, but language modeling requires object-level and relational infor- mation to be integrated as well as disentangled, and RA-based LMs have remained largely un- explored. BabyLM’s data-constrained training and comprehensive evaluation is an ideal testing ground for whether that data efficiency transfers. As a training intervention, we add a Next-Latent Prediction (NextLat; Teoh et al. 2026) objective that encourages hidden states to compress history incrementally into a dense belief state. Architecture is the dominant factor for structural linguistic generalization; the objective is secondary but still significant. DAT’s three relational attention types (full RA vs. the simpler RCA and DisRCA variants) are largely interchangeable at 10M words; full RA pulls ahead at 100M. We also introduce a novel symbol-retrieval mechanism (RoPE-based, as opposed to learned, relative symbols) that matches learned symbol libraries while adding no parameters. On the strict (100M- word) track, our best model ranks 6th of 55 overall and 3rd of 55 on the leaderboard’s NLP- task subset at the time of writing; our two strongest models outperform the GPT-2 base- line on most benchmarks, with one attaining the highest EWoK score among strict-track entries.</abstract>
    <identifier type="citekey">brasoveanu-etal-2026-relational</identifier>
    <location>
        <url>https://aclanthology.org/2026.babylm-main.21/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>260</start>
            <end>278</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Relational Attention for Data-Efficient Language Modeling
%A Brasoveanu, Adrian
%A Takmaz, Ece
%A Dotlacil, Jakub
%Y Choshen, Leshem
%Y Cotterell, Ryan
%Y Gul, Mustafa Omer
%Y Jumelet, Jaap
%Y Linzen, Tal
%Y Mueller, Aaron
%Y Salhan, Suchir
%Y Shah, Raj Sanjay
%Y Warstadt, Alex
%Y Wilcox, Ethan Gotlieb
%S Proceedings of the Second BabyLM Workshop
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-475-0
%F brasoveanu-etal-2026-relational
%X We present Relational BabyLM, a system submission to the BabyLM 2026 challenge that combines two cognitively motivated inductive biases in a single decoder-only Trans- former. Architecturally, we replace standard self-attention with a Dual Attention Trans- former (DAT), which separates the routing of object-level (“sensory”) lexical features from structural/relational information (Altabaa and Lafferty, 2025; Altabaa et al., 2024; Webb et al., 2024; Kerg et al., 2022; Webb et al., 2021). Relational attention (RA) disentangled from self-attention greatly increases data efficiency and out-of-training-sample generalization on purely relational tasks, but language modeling requires object-level and relational infor- mation to be integrated as well as disentangled, and RA-based LMs have remained largely un- explored. BabyLM’s data-constrained training and comprehensive evaluation is an ideal testing ground for whether that data efficiency transfers. As a training intervention, we add a Next-Latent Prediction (NextLat; Teoh et al. 2026) objective that encourages hidden states to compress history incrementally into a dense belief state. Architecture is the dominant factor for structural linguistic generalization; the objective is secondary but still significant. DAT’s three relational attention types (full RA vs. the simpler RCA and DisRCA variants) are largely interchangeable at 10M words; full RA pulls ahead at 100M. We also introduce a novel symbol-retrieval mechanism (RoPE-based, as opposed to learned, relative symbols) that matches learned symbol libraries while adding no parameters. On the strict (100M- word) track, our best model ranks 6th of 55 overall and 3rd of 55 on the leaderboard’s NLP- task subset at the time of writing; our two strongest models outperform the GPT-2 base- line on most benchmarks, with one attaining the highest EWoK score among strict-track entries.
%U https://aclanthology.org/2026.babylm-main.21/
%P 260-278

Markdown (Informal)

[Relational Attention for Data-Efficient Language Modeling](https://aclanthology.org/2026.babylm-main.21/) (Brasoveanu et al., BabyLM 2026)

• Relational Attention for Data-Efficient Language Modeling (Brasoveanu et al., BabyLM 2026)

ACL

• Adrian Brasoveanu, Ece Takmaz, and Jakub Dotlacil. 2026. Relational Attention for Data-Efficient Language Modeling. In Proceedings of the Second BabyLM Workshop, pages 260–278, Budapest, Hungary. Association for Computational Linguistics.