<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="botcazou-etal-2026-progress">
    <titleInfo>
        <title>Progress Ratio Embeddings: An Impatience Signal for Robust Length Control in Neural Text Generation</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Ivanhoé</namePart>
        <namePart type="family">Botcazou</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Tassadit</namePart>
        <namePart type="family">Amghar</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Sylvain</namePart>
        <namePart type="family">Lamprier</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Frédéric</namePart>
        <namePart type="family">Saubion</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-07</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics: ACL 2026</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Maria</namePart>
            <namePart type="family">Liakata</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Viviane</namePart>
            <namePart type="given">P</namePart>
            <namePart type="family">Moreira</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jiajun</namePart>
            <namePart type="family">Zhang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">David</namePart>
            <namePart type="family">Jurgens</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">San Diego, California, United States</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-395-1</identifier>
    </relatedItem>
    <abstract>Modern neural language models achieve high accuracy in text generation, yet precise control over generation length remains underdeveloped. In this paper, we first investigate a recent length control method based on Reverse Positional Embeddings (RPE) and show its limits when control is requested beyond the training distribution. In particular, using a discrete countdown signal tied to the absolute remaining token count leads to instability. To provide robust length control, we introduce Progress Ratio Embeddings (PRE), as continuous embeddings tied to a trigonometric impatience signal. PRE integrates seamlessly into standard Transformer architectures, providing stable length fidelity without degrading text accuracy under standard evaluation metrics. We further show that PRE generalizes well to unseen target lengths. Experiments on two widely used news-summarization benchmarks and a popular question generation dataset validate these findings.</abstract>
    <identifier type="citekey">botcazou-etal-2026-progress</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-acl.286/</url>
    </location>
    <part>
        <date>2026-07</date>
        <extent unit="page">
            <start>5776</start>
            <end>5789</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Progress Ratio Embeddings: An Impatience Signal for Robust Length Control in Neural Text Generation
%A Botcazou, Ivanhoé
%A Amghar, Tassadit
%A Lamprier, Sylvain
%A Saubion, Frédéric
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Findings of the Association for Computational Linguistics: ACL 2026
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-395-1
%F botcazou-etal-2026-progress
%X Modern neural language models achieve high accuracy in text generation, yet precise control over generation length remains underdeveloped. In this paper, we first investigate a recent length control method based on Reverse Positional Embeddings (RPE) and show its limits when control is requested beyond the training distribution. In particular, using a discrete countdown signal tied to the absolute remaining token count leads to instability. To provide robust length control, we introduce Progress Ratio Embeddings (PRE), as continuous embeddings tied to a trigonometric impatience signal. PRE integrates seamlessly into standard Transformer architectures, providing stable length fidelity without degrading text accuracy under standard evaluation metrics. We further show that PRE generalizes well to unseen target lengths. Experiments on two widely used news-summarization benchmarks and a popular question generation dataset validate these findings.
%U https://aclanthology.org/2026.findings-acl.286/
%P 5776-5789

Markdown (Informal)

[Progress Ratio Embeddings: An Impatience Signal for Robust Length Control in Neural Text Generation](https://aclanthology.org/2026.findings-acl.286/) (Botcazou et al., Findings 2026)

• Progress Ratio Embeddings: An Impatience Signal for Robust Length Control in Neural Text Generation (Botcazou et al., Findings 2026)

ACL

• Ivanhoé Botcazou, Tassadit Amghar, Sylvain Lamprier, and Frédéric Saubion. 2026. Progress Ratio Embeddings: An Impatience Signal for Robust Length Control in Neural Text Generation. In Findings of the Association for Computational Linguistics: ACL 2026, pages 5776–5789, San Diego, California, United States. Association for Computational Linguistics.