<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="belikova-etal-2026-detecting">
    <titleInfo>
        <title>Detecting Overflow in Compressed Token Representations for Retrieval-Augmented Generation</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Julia</namePart>
        <namePart type="family">Belikova</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Danila</namePart>
        <namePart type="family">Rozhevskii</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Dennis</namePart>
        <namePart type="family">Svirin</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Konstantin</namePart>
        <namePart type="family">Polev</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Alexander</namePart>
        <namePart type="family">Panchenko</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-03</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 4: Student Research Workshop)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Selene</namePart>
            <namePart type="family">Baez Santamaria</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Sai</namePart>
            <namePart type="given">Ashish</namePart>
            <namePart type="family">Somayajula</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Atsuki</namePart>
            <namePart type="family">Yamaguchi</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Rabat, Morocco</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-383-8</identifier>
    </relatedItem>
    <abstract>Efficient long-context processing remains a crucial challenge for contemporary large language models (LLMs), especially in resource-constrained environments. Soft compression architectures promise to extend effective context length by replacing long token sequences with smaller sets of learned compressed tokens. Yet, the limits of compressibility – and when compression begins to erase task-relevant content – remain underexplored. In this paper, we define token overflow as a regime in which compressed representations no longer contain sufficient information to answer a given query, and propose a methodology to characterize and detect it. In the xRAG soft-compression setting, we find that query-agnostic saturation statistics reliably separate compressed from uncompressed token representations, providing a practical tool for identifying compressed tokens but showing limited overflow detection capability. Lightweight probing classifiers over both query and context xRAG representations detect overflow with 0.72 AUC-ROC on average on HotpotQA, SQuADv2, and TriviaQA datasets, demonstrating that incorporating query information improves detection performance. These results advance from query-independent diagnostics to query-aware detectors, enabling low-cost pre-LLM gating to mitigate compression-induced errors.</abstract>
    <identifier type="citekey">belikova-etal-2026-detecting</identifier>
    <location>
        <url>https://aclanthology.org/2026.eacl-srw.59/</url>
    </location>
    <part>
        <date>2026-03</date>
        <extent unit="page">
            <start>797</start>
            <end>810</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Detecting Overflow in Compressed Token Representations for Retrieval-Augmented Generation
%A Belikova, Julia
%A Rozhevskii, Danila
%A Svirin, Dennis
%A Polev, Konstantin
%A Panchenko, Alexander
%Y Baez Santamaria, Selene
%Y Somayajula, Sai Ashish
%Y Yamaguchi, Atsuki
%S Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 4: Student Research Workshop)
%D 2026
%8 March
%I Association for Computational Linguistics
%C Rabat, Morocco
%@ 979-8-89176-383-8
%F belikova-etal-2026-detecting
%X Efficient long-context processing remains a crucial challenge for contemporary large language models (LLMs), especially in resource-constrained environments. Soft compression architectures promise to extend effective context length by replacing long token sequences with smaller sets of learned compressed tokens. Yet, the limits of compressibility – and when compression begins to erase task-relevant content – remain underexplored. In this paper, we define token overflow as a regime in which compressed representations no longer contain sufficient information to answer a given query, and propose a methodology to characterize and detect it. In the xRAG soft-compression setting, we find that query-agnostic saturation statistics reliably separate compressed from uncompressed token representations, providing a practical tool for identifying compressed tokens but showing limited overflow detection capability. Lightweight probing classifiers over both query and context xRAG representations detect overflow with 0.72 AUC-ROC on average on HotpotQA, SQuADv2, and TriviaQA datasets, demonstrating that incorporating query information improves detection performance. These results advance from query-independent diagnostics to query-aware detectors, enabling low-cost pre-LLM gating to mitigate compression-induced errors.
%U https://aclanthology.org/2026.eacl-srw.59/
%P 797-810

Markdown (Informal)

[Detecting Overflow in Compressed Token Representations for Retrieval-Augmented Generation](https://aclanthology.org/2026.eacl-srw.59/) (Belikova et al., EACL 2026)

• Detecting Overflow in Compressed Token Representations for Retrieval-Augmented Generation (Belikova et al., EACL 2026)

ACL

• Julia Belikova, Danila Rozhevskii, Dennis Svirin, Konstantin Polev, and Alexander Panchenko. 2026. Detecting Overflow in Compressed Token Representations for Retrieval-Augmented Generation. In Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 4: Student Research Workshop), pages 797–810, Rabat, Morocco. Association for Computational Linguistics.