@inproceedings{keles-deniz-2025-men,
title = "When Men Bite Dogs: Testing Good-Enough Parsing in {T}urkish with Humans and Large Language Models",
author = "Kele{\c{s}}, Onur and
Deniz, Nazik Dinctopal",
editor = "Kuribayashi, Tatsuki and
Rambelli, Giulia and
Takmaz, Ece and
Wicke, Philipp and
Li, Jixing and
Oh, Byung-Doh",
booktitle = "Proceedings of the Workshop on Cognitive Modeling and Computational Linguistics",
month = may,
year = "2025",
address = "Albuquerque, New Mexico, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.cmcl-1.26/",
doi = "10.18653/v1/2025.cmcl-1.26",
pages = "219--231",
ISBN = "979-8-89176-227-5",
abstract = "This paper investigates good-enough parsing in Turkish by comparing human self-paced reading performance to the surprisal and attention patterns of three Turkish Large Language Models (LLMs), GPT-2-Base, GPT-2-Large, and LLaMA-3. The results show that Turkish speakers rely on good-enough parsing for implausible but grammatically permissible sentences (e.g., interpreting sentences such as `the man bit the dog' as `the dog bit the man'). Although the smaller LLMs (e.g., GPT-2) were better predictors of human RTs, they seem to have relied more heavily on semantic plausibility than humans. Comparably, larger LLMs (e.g., LLaMA-3) tended to make more probabilistic parsing based on word order, exhibiting less \textit{good-enough} parsing behavior. Therefore, we conclude that LLMs take syntactic and semantic constraints into account when processing thematic roles, but not to the same extent as human parsers."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="keles-deniz-2025-men">
<titleInfo>
<title>When Men Bite Dogs: Testing Good-Enough Parsing in Turkish with Humans and Large Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Onur</namePart>
<namePart type="family">Keleş</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nazik</namePart>
<namePart type="given">Dinctopal</namePart>
<namePart type="family">Deniz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Workshop on Cognitive Modeling and Computational Linguistics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Tatsuki</namePart>
<namePart type="family">Kuribayashi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Giulia</namePart>
<namePart type="family">Rambelli</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ece</namePart>
<namePart type="family">Takmaz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Philipp</namePart>
<namePart type="family">Wicke</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jixing</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Byung-Doh</namePart>
<namePart type="family">Oh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Albuquerque, New Mexico, USA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-227-5</identifier>
</relatedItem>
<abstract>This paper investigates good-enough parsing in Turkish by comparing human self-paced reading performance to the surprisal and attention patterns of three Turkish Large Language Models (LLMs), GPT-2-Base, GPT-2-Large, and LLaMA-3. The results show that Turkish speakers rely on good-enough parsing for implausible but grammatically permissible sentences (e.g., interpreting sentences such as ‘the man bit the dog’ as ‘the dog bit the man’). Although the smaller LLMs (e.g., GPT-2) were better predictors of human RTs, they seem to have relied more heavily on semantic plausibility than humans. Comparably, larger LLMs (e.g., LLaMA-3) tended to make more probabilistic parsing based on word order, exhibiting less good-enough parsing behavior. Therefore, we conclude that LLMs take syntactic and semantic constraints into account when processing thematic roles, but not to the same extent as human parsers.</abstract>
<identifier type="citekey">keles-deniz-2025-men</identifier>
<identifier type="doi">10.18653/v1/2025.cmcl-1.26</identifier>
<location>
<url>https://aclanthology.org/2025.cmcl-1.26/</url>
</location>
<part>
<date>2025-05</date>
<extent unit="page">
<start>219</start>
<end>231</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T When Men Bite Dogs: Testing Good-Enough Parsing in Turkish with Humans and Large Language Models
%A Keleş, Onur
%A Deniz, Nazik Dinctopal
%Y Kuribayashi, Tatsuki
%Y Rambelli, Giulia
%Y Takmaz, Ece
%Y Wicke, Philipp
%Y Li, Jixing
%Y Oh, Byung-Doh
%S Proceedings of the Workshop on Cognitive Modeling and Computational Linguistics
%D 2025
%8 May
%I Association for Computational Linguistics
%C Albuquerque, New Mexico, USA
%@ 979-8-89176-227-5
%F keles-deniz-2025-men
%X This paper investigates good-enough parsing in Turkish by comparing human self-paced reading performance to the surprisal and attention patterns of three Turkish Large Language Models (LLMs), GPT-2-Base, GPT-2-Large, and LLaMA-3. The results show that Turkish speakers rely on good-enough parsing for implausible but grammatically permissible sentences (e.g., interpreting sentences such as ‘the man bit the dog’ as ‘the dog bit the man’). Although the smaller LLMs (e.g., GPT-2) were better predictors of human RTs, they seem to have relied more heavily on semantic plausibility than humans. Comparably, larger LLMs (e.g., LLaMA-3) tended to make more probabilistic parsing based on word order, exhibiting less good-enough parsing behavior. Therefore, we conclude that LLMs take syntactic and semantic constraints into account when processing thematic roles, but not to the same extent as human parsers.
%R 10.18653/v1/2025.cmcl-1.26
%U https://aclanthology.org/2025.cmcl-1.26/
%U https://doi.org/10.18653/v1/2025.cmcl-1.26
%P 219-231
Markdown (Informal)
[When Men Bite Dogs: Testing Good-Enough Parsing in Turkish with Humans and Large Language Models](https://aclanthology.org/2025.cmcl-1.26/) (Keleş & Deniz, CMCL 2025)
ACL