@inproceedings{kukic-etal-2026-sequence,
title = "Sequence Repetition Enhances Token Embeddings and Improves Sequence Labeling with Decoder-only Language Models",
author = "Kuki{\'c}, Matija Luka and
{\v{C}}uljak, Marko and
Duki{\'c}, David and
Tutek, Martin and
{\v{S}}najder, Jan",
editor = "Demberg, Vera and
Inui, Kentaro and
Marquez, Llu{\'i}s",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EACL} 2026",
month = mar,
year = "2026",
address = "Rabat, Morocco",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-eacl.339/",
pages = "6442--6456",
ISBN = "979-8-89176-386-9",
abstract = "Modern language models (LMs) are trained in an autoregressive manner, conditioned only on the prefix. In contrast, sequence labeling (SL) tasks assign labels to each individual input token, naturally benefiting from bidirectional context. This discrepancy has historically led SL to rely on inherently bidirectional encoder-only models. However, the rapid development of decoder-only models has raised the question of whether they can be adapted to SL. While causal mask removal has emerged as a viable technique for adapting decoder-only models to leverage the full context for SL, it requires considerable changes to the base model functionality. In this work, we explore sequence repetition (SR) as a less invasive alternative for enabling bidirectionality in decoder-only models. Through fine-tuning experiments, we show that SR inherently makes decoders bidirectional, improving the quality of token-level embeddings and surpassing encoders and unmasked decoders. Contrary to earlier claims, we find that increasing the number of repetitions does not degrade SL performance. Finally, we demonstrate that embeddings from intermediate layers are highly effective for SR, comparable to those from final layers, while being significantly more efficient to compute. Our findings underscore that SR alleviates the structural limitations of decoders, enabling more efficient and adaptable LMs and broadening their applicability to other token-level tasks."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kukic-etal-2026-sequence">
<titleInfo>
<title>Sequence Repetition Enhances Token Embeddings and Improves Sequence Labeling with Decoder-only Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Matija</namePart>
<namePart type="given">Luka</namePart>
<namePart type="family">Kukić</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marko</namePart>
<namePart type="family">Čuljak</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Dukić</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Martin</namePart>
<namePart type="family">Tutek</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jan</namePart>
<namePart type="family">Šnajder</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>Findings of the Association for Computational Linguistics: EACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Vera</namePart>
<namePart type="family">Demberg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kentaro</namePart>
<namePart type="family">Inui</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lluís</namePart>
<namePart type="family">Marquez</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">979-8-89176-386-9</identifier>
</relatedItem>
<abstract>Modern language models (LMs) are trained in an autoregressive manner, conditioned only on the prefix. In contrast, sequence labeling (SL) tasks assign labels to each individual input token, naturally benefiting from bidirectional context. This discrepancy has historically led SL to rely on inherently bidirectional encoder-only models. However, the rapid development of decoder-only models has raised the question of whether they can be adapted to SL. While causal mask removal has emerged as a viable technique for adapting decoder-only models to leverage the full context for SL, it requires considerable changes to the base model functionality. In this work, we explore sequence repetition (SR) as a less invasive alternative for enabling bidirectionality in decoder-only models. Through fine-tuning experiments, we show that SR inherently makes decoders bidirectional, improving the quality of token-level embeddings and surpassing encoders and unmasked decoders. Contrary to earlier claims, we find that increasing the number of repetitions does not degrade SL performance. Finally, we demonstrate that embeddings from intermediate layers are highly effective for SR, comparable to those from final layers, while being significantly more efficient to compute. Our findings underscore that SR alleviates the structural limitations of decoders, enabling more efficient and adaptable LMs and broadening their applicability to other token-level tasks.</abstract>
<identifier type="citekey">kukic-etal-2026-sequence</identifier>
<location>
<url>https://aclanthology.org/2026.findings-eacl.339/</url>
</location>
<part>
<date>2026-03</date>
<extent unit="page">
<start>6442</start>
<end>6456</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Sequence Repetition Enhances Token Embeddings and Improves Sequence Labeling with Decoder-only Language Models
%A Kukić, Matija Luka
%A Čuljak, Marko
%A Dukić, David
%A Tutek, Martin
%A Šnajder, Jan
%Y Demberg, Vera
%Y Inui, Kentaro
%Y Marquez, Lluís
%S Findings of the Association for Computational Linguistics: EACL 2026
%D 2026
%8 March
%I Association for Computational Linguistics
%C Rabat, Morocco
%@ 979-8-89176-386-9
%F kukic-etal-2026-sequence
%X Modern language models (LMs) are trained in an autoregressive manner, conditioned only on the prefix. In contrast, sequence labeling (SL) tasks assign labels to each individual input token, naturally benefiting from bidirectional context. This discrepancy has historically led SL to rely on inherently bidirectional encoder-only models. However, the rapid development of decoder-only models has raised the question of whether they can be adapted to SL. While causal mask removal has emerged as a viable technique for adapting decoder-only models to leverage the full context for SL, it requires considerable changes to the base model functionality. In this work, we explore sequence repetition (SR) as a less invasive alternative for enabling bidirectionality in decoder-only models. Through fine-tuning experiments, we show that SR inherently makes decoders bidirectional, improving the quality of token-level embeddings and surpassing encoders and unmasked decoders. Contrary to earlier claims, we find that increasing the number of repetitions does not degrade SL performance. Finally, we demonstrate that embeddings from intermediate layers are highly effective for SR, comparable to those from final layers, while being significantly more efficient to compute. Our findings underscore that SR alleviates the structural limitations of decoders, enabling more efficient and adaptable LMs and broadening their applicability to other token-level tasks.
%U https://aclanthology.org/2026.findings-eacl.339/
%P 6442-6456
Markdown (Informal)
[Sequence Repetition Enhances Token Embeddings and Improves Sequence Labeling with Decoder-only Language Models](https://aclanthology.org/2026.findings-eacl.339/) (Kukić et al., Findings 2026)
ACL