@inproceedings{wang-etal-2025-extracting,
title = "Extracting structure from an {LLM} - how to improve on surprisal-based models of Human Language Processing",
author = "Wang, Daphne P. and
Sadrzadeh, Mehrnoosh and
Stanojevi{\'c}, Milo{\v{s}} and
Chow, Wing-Yee and
Breheny, Richard",
editor = "Rambow, Owen and
Wanner, Leo and
Apidianaki, Marianna and
Al-Khalifa, Hend and
Eugenio, Barbara Di and
Schockaert, Steven",
booktitle = "Proceedings of the 31st International Conference on Computational Linguistics",
month = jan,
year = "2025",
address = "Abu Dhabi, UAE",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.coling-main.329/",
pages = "4938--4944",
abstract = "Prediction and reanalysis are considered two key processes that underly humans' capacity to comprehend language in real time. Computational models capture it using Large Language Models (LLMs) and a statistical measure known as {\textquoteleft}surprisal'. Despite successes of LLMs, surprisal-based models face challenges when it comes to sentences requiring reanalysis due to pervasive temporary structural ambiguities, such as garden path sentences. We ask whether structural information can be extracted from LLM`s and develop a model that integrates it with their learnt statistics. When applied to a dataset of garden path sentences, the model achieved a significantly higher correlation with human reading times than surprisal. It also provided a better prediction of the garden path effect and could distinguish between sentence types with different levels of difficulty."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2025-extracting">
<titleInfo>
<title>Extracting structure from an LLM - how to improve on surprisal-based models of Human Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Daphne</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mehrnoosh</namePart>
<namePart type="family">Sadrzadeh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Miloš</namePart>
<namePart type="family">Stanojević</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wing-Yee</namePart>
<namePart type="family">Chow</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Richard</namePart>
<namePart type="family">Breheny</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-01</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 31st International Conference on Computational Linguistics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Owen</namePart>
<namePart type="family">Rambow</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Leo</namePart>
<namePart type="family">Wanner</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marianna</namePart>
<namePart type="family">Apidianaki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hend</namePart>
<namePart type="family">Al-Khalifa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Barbara</namePart>
<namePart type="given">Di</namePart>
<namePart type="family">Eugenio</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Steven</namePart>
<namePart type="family">Schockaert</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Abu Dhabi, UAE</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Prediction and reanalysis are considered two key processes that underly humans’ capacity to comprehend language in real time. Computational models capture it using Large Language Models (LLMs) and a statistical measure known as ‘surprisal’. Despite successes of LLMs, surprisal-based models face challenges when it comes to sentences requiring reanalysis due to pervasive temporary structural ambiguities, such as garden path sentences. We ask whether structural information can be extracted from LLM‘s and develop a model that integrates it with their learnt statistics. When applied to a dataset of garden path sentences, the model achieved a significantly higher correlation with human reading times than surprisal. It also provided a better prediction of the garden path effect and could distinguish between sentence types with different levels of difficulty.</abstract>
<identifier type="citekey">wang-etal-2025-extracting</identifier>
<location>
<url>https://aclanthology.org/2025.coling-main.329/</url>
</location>
<part>
<date>2025-01</date>
<extent unit="page">
<start>4938</start>
<end>4944</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Extracting structure from an LLM - how to improve on surprisal-based models of Human Language Processing
%A Wang, Daphne P.
%A Sadrzadeh, Mehrnoosh
%A Stanojević, Miloš
%A Chow, Wing-Yee
%A Breheny, Richard
%Y Rambow, Owen
%Y Wanner, Leo
%Y Apidianaki, Marianna
%Y Al-Khalifa, Hend
%Y Eugenio, Barbara Di
%Y Schockaert, Steven
%S Proceedings of the 31st International Conference on Computational Linguistics
%D 2025
%8 January
%I Association for Computational Linguistics
%C Abu Dhabi, UAE
%F wang-etal-2025-extracting
%X Prediction and reanalysis are considered two key processes that underly humans’ capacity to comprehend language in real time. Computational models capture it using Large Language Models (LLMs) and a statistical measure known as ‘surprisal’. Despite successes of LLMs, surprisal-based models face challenges when it comes to sentences requiring reanalysis due to pervasive temporary structural ambiguities, such as garden path sentences. We ask whether structural information can be extracted from LLM‘s and develop a model that integrates it with their learnt statistics. When applied to a dataset of garden path sentences, the model achieved a significantly higher correlation with human reading times than surprisal. It also provided a better prediction of the garden path effect and could distinguish between sentence types with different levels of difficulty.
%U https://aclanthology.org/2025.coling-main.329/
%P 4938-4944
Markdown (Informal)
[Extracting structure from an LLM - how to improve on surprisal-based models of Human Language Processing](https://aclanthology.org/2025.coling-main.329/) (Wang et al., COLING 2025)
ACL