@inproceedings{chen-etal-2026-optimizing,
title = "Optimizing Packing and Shuffling Strategies for Enhanced Performance in Generative Language Models",
author = "Chen, Yanbing and
Wang, Ruilin and
Yang, Zihao and
Jiang, Lavender Yao and
Oermann, Eric Karl",
editor = "T.Y.S.S., Santosh and
Rodriguez, Juan Diego and
de Gibert, Ona",
booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics ({ACL} 2026)",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.acl-srw.124/",
pages = "1399--1416",
ISBN = "979-8-89176-393-7",
abstract = "Packing and shuffling tokens is a common practice in training auto-regressive language models to prevent overfitting and improve efficiency. Documents are typically concatenated to chunks of maximum sequence length (MSL) and shuffled in chunks of tokens (atom-size chunk), possibly breaking context within documents. An alternative approach is padding, which only includes one document per chunk. To optimize both packing strategies (concatenation vs padding), we explored the optimal atom size for shuffling and compared performance and efficiency. We found that in the most common setup (where average document length is greater than MSL), matching atom size to MSL yields the lowest perplexity, controlling for dataset. Also, padding yields lower final perplexity than concatenation at the cost of lower efficiency. This trade-off informs the choice of shuffling and packing methods in training LMs."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="chen-etal-2026-optimizing">
<titleInfo>
<title>Optimizing Packing and Shuffling Strategies for Enhanced Performance in Generative Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yanbing</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruilin</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zihao</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lavender</namePart>
<namePart type="given">Yao</namePart>
<namePart type="family">Jiang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Eric</namePart>
<namePart type="given">Karl</namePart>
<namePart type="family">Oermann</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>Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Santosh</namePart>
<namePart type="family">T.Y.S.S.</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Juan</namePart>
<namePart type="given">Diego</namePart>
<namePart type="family">Rodriguez</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ona</namePart>
<namePart type="family">de Gibert</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">979-8-89176-393-7</identifier>
</relatedItem>
<abstract>Packing and shuffling tokens is a common practice in training auto-regressive language models to prevent overfitting and improve efficiency. Documents are typically concatenated to chunks of maximum sequence length (MSL) and shuffled in chunks of tokens (atom-size chunk), possibly breaking context within documents. An alternative approach is padding, which only includes one document per chunk. To optimize both packing strategies (concatenation vs padding), we explored the optimal atom size for shuffling and compared performance and efficiency. We found that in the most common setup (where average document length is greater than MSL), matching atom size to MSL yields the lowest perplexity, controlling for dataset. Also, padding yields lower final perplexity than concatenation at the cost of lower efficiency. This trade-off informs the choice of shuffling and packing methods in training LMs.</abstract>
<identifier type="citekey">chen-etal-2026-optimizing</identifier>
<location>
<url>https://aclanthology.org/2026.acl-srw.124/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>1399</start>
<end>1416</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Optimizing Packing and Shuffling Strategies for Enhanced Performance in Generative Language Models
%A Chen, Yanbing
%A Wang, Ruilin
%A Yang, Zihao
%A Jiang, Lavender Yao
%A Oermann, Eric Karl
%Y T.Y.S.S., Santosh
%Y Rodriguez, Juan Diego
%Y de Gibert, Ona
%S Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-393-7
%F chen-etal-2026-optimizing
%X Packing and shuffling tokens is a common practice in training auto-regressive language models to prevent overfitting and improve efficiency. Documents are typically concatenated to chunks of maximum sequence length (MSL) and shuffled in chunks of tokens (atom-size chunk), possibly breaking context within documents. An alternative approach is padding, which only includes one document per chunk. To optimize both packing strategies (concatenation vs padding), we explored the optimal atom size for shuffling and compared performance and efficiency. We found that in the most common setup (where average document length is greater than MSL), matching atom size to MSL yields the lowest perplexity, controlling for dataset. Also, padding yields lower final perplexity than concatenation at the cost of lower efficiency. This trade-off informs the choice of shuffling and packing methods in training LMs.
%U https://aclanthology.org/2026.acl-srw.124/
%P 1399-1416
Markdown (Informal)
[Optimizing Packing and Shuffling Strategies for Enhanced Performance in Generative Language Models](https://aclanthology.org/2026.acl-srw.124/) (Chen et al., ACL 2026)
ACL