@inproceedings{payne-kodner-2025-lemmas,
title = "Lemmas Matter, But Not Like That: Predictors of Lemma-Based Generalization in Morphological Inflection",
author = "Payne, Sarah and
Kodner, Jordan",
editor = "Che, Wanxiang and
Nabende, Joyce and
Shutova, Ekaterina and
Pilehvar, Mohammad Taher",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2025",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-acl.1296/",
doi = "10.18653/v1/2025.findings-acl.1296",
pages = "25270--25286",
ISBN = "979-8-89176-256-5",
abstract = "Recent work has suggested that overlap {--}whether a given lemma or feature set is attested independently in train {--} drives model performance on morphological inflection tasks. The impact of lemma overlap, however, is debated, with recent work reporting accuracy drops from 0{\%} to 30{\%} between seen and unseen test lemmas. In this paper, we introduce a novel splitting algorithm designed to investigate predictors of accuracy on seen and unseen lemmas. We find only an 11{\%} average drop from seen to unseen test lemmas, but show that the number of lemmas in train has a much stronger effect on accuracy on unseen than seen lemmas. We also show that the previously reported 30{\%} drop is inflated due to the introduction of a near-30{\%} drop in the number of training lemmas from the original splits to their novel splits."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="payne-kodner-2025-lemmas">
<titleInfo>
<title>Lemmas Matter, But Not Like That: Predictors of Lemma-Based Generalization in Morphological Inflection</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sarah</namePart>
<namePart type="family">Payne</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jordan</namePart>
<namePart type="family">Kodner</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2025</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wanxiang</namePart>
<namePart type="family">Che</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joyce</namePart>
<namePart type="family">Nabende</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Shutova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohammad</namePart>
<namePart type="given">Taher</namePart>
<namePart type="family">Pilehvar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Vienna, Austria</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-256-5</identifier>
</relatedItem>
<abstract>Recent work has suggested that overlap –whether a given lemma or feature set is attested independently in train – drives model performance on morphological inflection tasks. The impact of lemma overlap, however, is debated, with recent work reporting accuracy drops from 0% to 30% between seen and unseen test lemmas. In this paper, we introduce a novel splitting algorithm designed to investigate predictors of accuracy on seen and unseen lemmas. We find only an 11% average drop from seen to unseen test lemmas, but show that the number of lemmas in train has a much stronger effect on accuracy on unseen than seen lemmas. We also show that the previously reported 30% drop is inflated due to the introduction of a near-30% drop in the number of training lemmas from the original splits to their novel splits.</abstract>
<identifier type="citekey">payne-kodner-2025-lemmas</identifier>
<identifier type="doi">10.18653/v1/2025.findings-acl.1296</identifier>
<location>
<url>https://aclanthology.org/2025.findings-acl.1296/</url>
</location>
<part>
<date>2025-07</date>
<extent unit="page">
<start>25270</start>
<end>25286</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Lemmas Matter, But Not Like That: Predictors of Lemma-Based Generalization in Morphological Inflection
%A Payne, Sarah
%A Kodner, Jordan
%Y Che, Wanxiang
%Y Nabende, Joyce
%Y Shutova, Ekaterina
%Y Pilehvar, Mohammad Taher
%S Findings of the Association for Computational Linguistics: ACL 2025
%D 2025
%8 July
%I Association for Computational Linguistics
%C Vienna, Austria
%@ 979-8-89176-256-5
%F payne-kodner-2025-lemmas
%X Recent work has suggested that overlap –whether a given lemma or feature set is attested independently in train – drives model performance on morphological inflection tasks. The impact of lemma overlap, however, is debated, with recent work reporting accuracy drops from 0% to 30% between seen and unseen test lemmas. In this paper, we introduce a novel splitting algorithm designed to investigate predictors of accuracy on seen and unseen lemmas. We find only an 11% average drop from seen to unseen test lemmas, but show that the number of lemmas in train has a much stronger effect on accuracy on unseen than seen lemmas. We also show that the previously reported 30% drop is inflated due to the introduction of a near-30% drop in the number of training lemmas from the original splits to their novel splits.
%R 10.18653/v1/2025.findings-acl.1296
%U https://aclanthology.org/2025.findings-acl.1296/
%U https://doi.org/10.18653/v1/2025.findings-acl.1296
%P 25270-25286
Markdown (Informal)
[Lemmas Matter, But Not Like That: Predictors of Lemma-Based Generalization in Morphological Inflection](https://aclanthology.org/2025.findings-acl.1296/) (Payne & Kodner, Findings 2025)
ACL