@inproceedings{fusco-etal-2025-linguistic,
title = "Linguistic Units as Tokens: Intrinsic and Extrinsic Evaluation with {B}aby{LM}",
author = "Fusco, Achille and
Bianchessi, Maria Letizia Piccini and
Sgrizzi, Tommaso and
Zanollo, Asya and
Chesi, Cristiano",
editor = "Charpentier, Lucas and
Choshen, Leshem and
Cotterell, Ryan and
Gul, Mustafa Omer and
Hu, Michael Y. and
Liu, Jing and
Jumelet, Jaap and
Linzen, Tal and
Mueller, Aaron and
Ross, Candace and
Shah, Raj Sanjay and
Warstadt, Alex and
Wilcox, Ethan Gotlieb and
Williams, Adina",
booktitle = "Proceedings of the First BabyLM Workshop",
month = nov,
year = "2025",
address = "Suzhou, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.babylm-main.35/",
pages = "496--507",
ISBN = "TODO",
abstract = "Tokenization is often treated as a preprocessing step, yet in data-limited settings it directly shapes what a model can learn. We compare four segmentation strategies in the BabyLM Challenge: frequency-based BPE, morphology-aware MorPiece and ParadigmFinder, and syllable-based SylliTok. Evaluation combines two perspectives. First, an intrinsic test on the SIGMORPHON 2022 segmentation benchmark, adapted to English, measures how closely each tokenizer aligns with morpheme boundaries. Second, extrinsic tests train GPT-2 on the 10M BabyLM corpus and evaluate on the 2025 benchmark. No single tokenizer dominates. BPE remains strong on syntax-heavy tasks. ParadigmFinder excels in semantic composition and age-of-acquisition alignment. MorPiece shows advantages in discourse tracking. Morphology-aware tokenizers achieve the best intrinsic segmentation scores, and these gains translate into more robust generalisation in comprehension tasks. These results highlight tokenization as a core modeling decision, with direct consequences for compression, morphology, and the path to humanlike learning."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="fusco-etal-2025-linguistic">
<titleInfo>
<title>Linguistic Units as Tokens: Intrinsic and Extrinsic Evaluation with BabyLM</title>
</titleInfo>
<name type="personal">
<namePart type="given">Achille</namePart>
<namePart type="family">Fusco</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="given">Letizia</namePart>
<namePart type="given">Piccini</namePart>
<namePart type="family">Bianchessi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tommaso</namePart>
<namePart type="family">Sgrizzi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Asya</namePart>
<namePart type="family">Zanollo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cristiano</namePart>
<namePart type="family">Chesi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the First BabyLM Workshop</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lucas</namePart>
<namePart type="family">Charpentier</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Leshem</namePart>
<namePart type="family">Choshen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ryan</namePart>
<namePart type="family">Cotterell</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mustafa</namePart>
<namePart type="given">Omer</namePart>
<namePart type="family">Gul</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Michael</namePart>
<namePart type="given">Y</namePart>
<namePart type="family">Hu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jing</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jaap</namePart>
<namePart type="family">Jumelet</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tal</namePart>
<namePart type="family">Linzen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aaron</namePart>
<namePart type="family">Mueller</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Candace</namePart>
<namePart type="family">Ross</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Raj</namePart>
<namePart type="given">Sanjay</namePart>
<namePart type="family">Shah</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alex</namePart>
<namePart type="family">Warstadt</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ethan</namePart>
<namePart type="given">Gotlieb</namePart>
<namePart type="family">Wilcox</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Adina</namePart>
<namePart type="family">Williams</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Suzhou, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">TODO</identifier>
</relatedItem>
<abstract>Tokenization is often treated as a preprocessing step, yet in data-limited settings it directly shapes what a model can learn. We compare four segmentation strategies in the BabyLM Challenge: frequency-based BPE, morphology-aware MorPiece and ParadigmFinder, and syllable-based SylliTok. Evaluation combines two perspectives. First, an intrinsic test on the SIGMORPHON 2022 segmentation benchmark, adapted to English, measures how closely each tokenizer aligns with morpheme boundaries. Second, extrinsic tests train GPT-2 on the 10M BabyLM corpus and evaluate on the 2025 benchmark. No single tokenizer dominates. BPE remains strong on syntax-heavy tasks. ParadigmFinder excels in semantic composition and age-of-acquisition alignment. MorPiece shows advantages in discourse tracking. Morphology-aware tokenizers achieve the best intrinsic segmentation scores, and these gains translate into more robust generalisation in comprehension tasks. These results highlight tokenization as a core modeling decision, with direct consequences for compression, morphology, and the path to humanlike learning.</abstract>
<identifier type="citekey">fusco-etal-2025-linguistic</identifier>
<location>
<url>https://aclanthology.org/2025.babylm-main.35/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>496</start>
<end>507</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Linguistic Units as Tokens: Intrinsic and Extrinsic Evaluation with BabyLM
%A Fusco, Achille
%A Bianchessi, Maria Letizia Piccini
%A Sgrizzi, Tommaso
%A Zanollo, Asya
%A Chesi, Cristiano
%Y Charpentier, Lucas
%Y Choshen, Leshem
%Y Cotterell, Ryan
%Y Gul, Mustafa Omer
%Y Hu, Michael Y.
%Y Liu, Jing
%Y Jumelet, Jaap
%Y Linzen, Tal
%Y Mueller, Aaron
%Y Ross, Candace
%Y Shah, Raj Sanjay
%Y Warstadt, Alex
%Y Wilcox, Ethan Gotlieb
%Y Williams, Adina
%S Proceedings of the First BabyLM Workshop
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ TODO
%F fusco-etal-2025-linguistic
%X Tokenization is often treated as a preprocessing step, yet in data-limited settings it directly shapes what a model can learn. We compare four segmentation strategies in the BabyLM Challenge: frequency-based BPE, morphology-aware MorPiece and ParadigmFinder, and syllable-based SylliTok. Evaluation combines two perspectives. First, an intrinsic test on the SIGMORPHON 2022 segmentation benchmark, adapted to English, measures how closely each tokenizer aligns with morpheme boundaries. Second, extrinsic tests train GPT-2 on the 10M BabyLM corpus and evaluate on the 2025 benchmark. No single tokenizer dominates. BPE remains strong on syntax-heavy tasks. ParadigmFinder excels in semantic composition and age-of-acquisition alignment. MorPiece shows advantages in discourse tracking. Morphology-aware tokenizers achieve the best intrinsic segmentation scores, and these gains translate into more robust generalisation in comprehension tasks. These results highlight tokenization as a core modeling decision, with direct consequences for compression, morphology, and the path to humanlike learning.
%U https://aclanthology.org/2025.babylm-main.35/
%P 496-507
Markdown (Informal)
[Linguistic Units as Tokens: Intrinsic and Extrinsic Evaluation with BabyLM](https://aclanthology.org/2025.babylm-main.35/) (Fusco et al., BabyLM 2025)
ACL