@inproceedings{zehe-etal-2025-assessing,
title = "Assessing the State of the Art in Scene Segmentation",
author = "Zehe, Albin and
Fischer, Elisabeth and
Hotho, Andreas",
editor = "Chiruzzo, Luis and
Ritter, Alan and
Wang, Lu",
booktitle = "Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)",
month = apr,
year = "2025",
address = "Albuquerque, New Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.naacl-long.500/",
doi = "10.18653/v1/2025.naacl-long.500",
pages = "9922--9941",
ISBN = "979-8-89176-189-6",
abstract = "The detection of scenes in literary texts is a recently introduced segmentation task in computational literary studies. Its goal is to partition a fictional text into segments that are coherent across the dimensions time, space, action and character constellation. This task is very challenging for automatic methods, since it requires a high-level understanding of the text. In this paper, we provide a thorough analysis of the State of the Art and challenges in this task, identifying and solving a problem in the training procedure for previous approaches, analysing the generalisation capabilities of the models and comparing the BERT-based SotA to current Llama models, as well as providing an analysis of what causes errors in the models. Our change in training procedure provides a significant increase in performance. We find that Llama-based models are more robust to different types of texts, while their overall performance is slightly worse than that of BERT-based models."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zehe-etal-2025-assessing">
<titleInfo>
<title>Assessing the State of the Art in Scene Segmentation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Albin</namePart>
<namePart type="family">Zehe</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Elisabeth</namePart>
<namePart type="family">Fischer</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andreas</namePart>
<namePart type="family">Hotho</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-04</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Luis</namePart>
<namePart type="family">Chiruzzo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alan</namePart>
<namePart type="family">Ritter</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lu</namePart>
<namePart type="family">Wang</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</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-189-6</identifier>
</relatedItem>
<abstract>The detection of scenes in literary texts is a recently introduced segmentation task in computational literary studies. Its goal is to partition a fictional text into segments that are coherent across the dimensions time, space, action and character constellation. This task is very challenging for automatic methods, since it requires a high-level understanding of the text. In this paper, we provide a thorough analysis of the State of the Art and challenges in this task, identifying and solving a problem in the training procedure for previous approaches, analysing the generalisation capabilities of the models and comparing the BERT-based SotA to current Llama models, as well as providing an analysis of what causes errors in the models. Our change in training procedure provides a significant increase in performance. We find that Llama-based models are more robust to different types of texts, while their overall performance is slightly worse than that of BERT-based models.</abstract>
<identifier type="citekey">zehe-etal-2025-assessing</identifier>
<identifier type="doi">10.18653/v1/2025.naacl-long.500</identifier>
<location>
<url>https://aclanthology.org/2025.naacl-long.500/</url>
</location>
<part>
<date>2025-04</date>
<extent unit="page">
<start>9922</start>
<end>9941</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Assessing the State of the Art in Scene Segmentation
%A Zehe, Albin
%A Fischer, Elisabeth
%A Hotho, Andreas
%Y Chiruzzo, Luis
%Y Ritter, Alan
%Y Wang, Lu
%S Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)
%D 2025
%8 April
%I Association for Computational Linguistics
%C Albuquerque, New Mexico
%@ 979-8-89176-189-6
%F zehe-etal-2025-assessing
%X The detection of scenes in literary texts is a recently introduced segmentation task in computational literary studies. Its goal is to partition a fictional text into segments that are coherent across the dimensions time, space, action and character constellation. This task is very challenging for automatic methods, since it requires a high-level understanding of the text. In this paper, we provide a thorough analysis of the State of the Art and challenges in this task, identifying and solving a problem in the training procedure for previous approaches, analysing the generalisation capabilities of the models and comparing the BERT-based SotA to current Llama models, as well as providing an analysis of what causes errors in the models. Our change in training procedure provides a significant increase in performance. We find that Llama-based models are more robust to different types of texts, while their overall performance is slightly worse than that of BERT-based models.
%R 10.18653/v1/2025.naacl-long.500
%U https://aclanthology.org/2025.naacl-long.500/
%U https://doi.org/10.18653/v1/2025.naacl-long.500
%P 9922-9941
Markdown (Informal)
[Assessing the State of the Art in Scene Segmentation](https://aclanthology.org/2025.naacl-long.500/) (Zehe et al., NAACL 2025)
ACL
- Albin Zehe, Elisabeth Fischer, and Andreas Hotho. 2025. Assessing the State of the Art in Scene Segmentation. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), pages 9922–9941, Albuquerque, New Mexico. Association for Computational Linguistics.