<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="stearns-etal-2026-cross">
    <titleInfo>
        <title>Cross-Corpus CEFR Classification through Artificial Learners Perplexities</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Bernardo</namePart>
        <namePart type="family">Stearns</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">John</namePart>
        <namePart type="given">P</namePart>
        <namePart type="family">McCrae</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Thomas</namePart>
        <namePart type="family">Gaillat</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-05</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the Fifteenth Language Resources and Evaluation Conference</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Stelios</namePart>
            <namePart type="family">Piperidis</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Núria</namePart>
            <namePart type="family">Bel</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Henk</namePart>
            <namePart type="family">van den Heuvel</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Nancy</namePart>
            <namePart type="family">Ide</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Simon</namePart>
            <namePart type="family">Krek</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Antonio</namePart>
            <namePart type="family">Toral</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>ELRA Language Resource Association</publisher>
            <place>
                <placeTerm type="text">Palma de Mallorca, Spain</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>The complexity of neural methods for automatic proficiency assessment often sacrifices interpretability and robustness. This paper presents a competitive alternative for CEFR classification using optimized statistical models with a novel perplexity-based feature engineering pipeline. We introduce LLM-derived perplexity features as a proxy for how unexpected a learner’s word choices are: native model perplexity measures unexpectedness relative to native language use, while Artificial Learner model perplexity quantifies relative to a specific proficiency level. While recent work favors end-to-end neural architectures, we demonstrate that traditional pipelines enhanced with these interpretable perplexity features can achieve comparable performance on established benchmarks. We evaluate two transfer scenarios: zero-shot (trained on EFCAMDAT, tested on external corpora) and 90-10 split (same features, in-domain classifier training). On KUPA-KEYS, perplexity features achieve RMSE 0.707 (zero-shot) and 0.660 (90-10 split), outperforming fine-tuned BERT and prompt-based LLMs. On CELVA-SP, zero-shot perplexity shows limited generalization (RMSE 1.437 vs. LLM’s 1.016), but statistical models close this gap in the 90-10 split (RMSE 0.872). Across all three evaluation datasets, perplexity-based models achieve the best average macro F1 in the 90-10 split (0.446 vs. 0.287 for BERT and 0.175 for prompting), demonstrating that interpretable features paired with domain-adapted classifiers provide the most robust cross-domain representations. We contribute: (1) state-of-the-art KUPA-KEYS results with interpretable models, (2) the first comprehensive CELVA-SP benchmark, and (3) evidence that feature-level transfer outperforms both end-to-end fine-tuning and zero-shot prompting.</abstract>
    <identifier type="citekey">stearns-etal-2026-cross</identifier>
    <identifier type="doi">10.63317/2bvpcczgao2t</identifier>
    <location>
        <url>https://aclanthology.org/2026.lrec-1.62/</url>
    </location>
    <part>
        <date>2026-05</date>
        <extent unit="page">
            <start>828</start>
            <end>837</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Cross-Corpus CEFR Classification through Artificial Learners Perplexities
%A Stearns, Bernardo
%A McCrae, John P.
%A Gaillat, Thomas
%Y Piperidis, Stelios
%Y Bel, Núria
%Y van den Heuvel, Henk
%Y Ide, Nancy
%Y Krek, Simon
%Y Toral, Antonio
%S Proceedings of the Fifteenth Language Resources and Evaluation Conference
%D 2026
%8 May
%I ELRA Language Resource Association
%C Palma de Mallorca, Spain
%F stearns-etal-2026-cross
%X The complexity of neural methods for automatic proficiency assessment often sacrifices interpretability and robustness. This paper presents a competitive alternative for CEFR classification using optimized statistical models with a novel perplexity-based feature engineering pipeline. We introduce LLM-derived perplexity features as a proxy for how unexpected a learner’s word choices are: native model perplexity measures unexpectedness relative to native language use, while Artificial Learner model perplexity quantifies relative to a specific proficiency level. While recent work favors end-to-end neural architectures, we demonstrate that traditional pipelines enhanced with these interpretable perplexity features can achieve comparable performance on established benchmarks. We evaluate two transfer scenarios: zero-shot (trained on EFCAMDAT, tested on external corpora) and 90-10 split (same features, in-domain classifier training). On KUPA-KEYS, perplexity features achieve RMSE 0.707 (zero-shot) and 0.660 (90-10 split), outperforming fine-tuned BERT and prompt-based LLMs. On CELVA-SP, zero-shot perplexity shows limited generalization (RMSE 1.437 vs. LLM’s 1.016), but statistical models close this gap in the 90-10 split (RMSE 0.872). Across all three evaluation datasets, perplexity-based models achieve the best average macro F1 in the 90-10 split (0.446 vs. 0.287 for BERT and 0.175 for prompting), demonstrating that interpretable features paired with domain-adapted classifiers provide the most robust cross-domain representations. We contribute: (1) state-of-the-art KUPA-KEYS results with interpretable models, (2) the first comprehensive CELVA-SP benchmark, and (3) evidence that feature-level transfer outperforms both end-to-end fine-tuning and zero-shot prompting.
%R 10.63317/2bvpcczgao2t
%U https://aclanthology.org/2026.lrec-1.62/
%U https://doi.org/10.63317/2bvpcczgao2t
%P 828-837

Markdown (Informal)

[Cross-Corpus CEFR Classification through Artificial Learners Perplexities](https://aclanthology.org/2026.lrec-1.62/) (Stearns et al., LREC 2026)

• Cross-Corpus CEFR Classification through Artificial Learners Perplexities (Stearns et al., LREC 2026)

ACL

• Bernardo Stearns, John P. McCrae, and Thomas Gaillat. 2026. Cross-Corpus CEFR Classification through Artificial Learners Perplexities. In Proceedings of the Fifteenth Language Resources and Evaluation Conference, pages 828–837, Palma de Mallorca, Spain. ELRA Language Resource Association.