April 10th, 2025 - Lucca, Italy

Text2Story 2025

Eighth International Workshop on Narrative Extraction from Texts
held in conjunction with the 47th European Conference on Information Retrieval

Call for papers

Overview

For seven years, the Text2Story Workshop series has fostered a vibrant community dedicated to understanding narrative structure in text, resulting in significant contributions to the field and developing a shared understanding of the challenges in this domain. While traditional methods have yielded valuable insights, the advent of Transformers and LLMs have ignited a new wave of interest in narrative understanding. In the eighth edition of the Text2Story workshop, we propose to go deeper into the role of LLMs in narrative understanding exploring the issues involved in using LLMs to unravel narrative structures, while also examining the characteristics of narratives generated by LLMs. By fostering dialogue on these emerging areas, we aim to identify the wide-ranging issues related to the narrative extraction task and continue the workshop's tradition of driving innovation in narrative understanding research.

Call for papers

Research works submitted to the workshop should advance the scientific understanding of all aspects of narrative extraction from texts. This includes, but is not limited to, topics such as narrative information extraction, formal representation of narratives, narrative analysis and generation, development of datasets and evaluation protocols, as well as ethics and bias in narratives, and narrative applications. We encourage the submission of high-quality and original submissions covering the following topics and contributions focused on low and medium-resource languages.

    Information Extraction Aspects

  • Temporal Relation Identification
  • Temporal Reasoning and Ordering of Events
  • Causal Relation Extraction and Arrangement
  • Big Data Applied to Narrative Extraction
  • Narrative Representation

  • Annotation protocols
  • Narrative Representation Models
  • Lexical, Syntactic, and Semantic Ambiguity in Narrative Representation
  • Narrative Analysis and Generation

  • Argumentation Analysis
  • Language Models and Transfer Learning in Narrative Analysis
  • Narrative Analysis in Low-resource Languages
  • Multilinguality: Multilingual and Cross-lingual Narrative Analysis
  • Comprehension of Generated Narratives
  • Story Evolution and Shift Detection
  • Automatic Timeline Generation
  • Datasets and Evaluation Protocol

  • Evaluation Methodologies for Narrative Extraction
  • Annotated datasets
  • Narrative Resources
  • Ethics and Bias in Narratives

  • Bias Detection and Removal in Generated Stories
  • Ethical and Fair Narrative Generation
  • Misinformation and Fact Checking
  • Narrative Applications

  • Narrative-focused Search in Text Collections
  • Narrative Summarization
  • Narrative Q&A
  • Multi-modal Narrative Summarization
  • Sentiment and Opinion Detection in Narratives
  • Social Media Narratives
  • Narrative Simplification
  • Personalization and Recommendation of Narratives
  • Storyline Visualization

Objectives

Overall, the workshop has the following main objectives: (1) raise awareness within the Information Retrieval (IR) community regarding the challenges posed by narrative extraction and comprehension; (2) bridge the gap and foster connections between academic research, practitioners, and industrial applications; (3) discuss new methods, recent advances, and emerging challenges; (4) share experiences from research projects, case studies, and scientific outcomes structured around fundamental research questions related to narrative understanding; (5) identify dimensions that might be influenced by the automation of the narrative process; (6) highlight tested hypotheses that did not result in the expected outcomes

Important Dates

  • February 7th, 2025
    Submission Deadline
  • March 3rd, 2025
    Acceptance Notification
  • March 17th, 2025
    Camera-ready copies
  • April 10th, 2025
    Workshop

Submissions

We expect contributions from researchers on all aspects of narrative extraction, representation, analysis, and generation. This includes the extraction and formal representation of events, their temporal and causal relationships, and methods for temporal reasoning and ordering. Submissions focusing on narrative comprehension, such as the analysis of generated narratives, are also highly encouraged. Additionally, we welcome innovative approaches to presenting narrative information, including automatic timeline generation, multi-modal narrative summarization, and narrative visualization. Research addressing misinformation and the verification of extracted facts, evaluation methodologies, and the development of annotated datasets, annotation schemas, and evaluation metrics is particularly valued. Finally, we are especially interested in submissions that focus on low and medium-resource languages, as well as multilingual and cross-lingual narrative analysis.

Building on these themes, several pressing questions emerge within the field, offering valuable guidance for authors in shaping their submissions.How can we better integrate multimodal content - combining text, images, videos, and audio - into cohesive narratives? What strategies can reliably extract or generate accurate narratives from large, multi-genre, and multi-lingual datasets? How can systems dynamically adapt to real-time shifts in narratives as the volume of generated content grows? What methodologies can effectively annotate data and evaluate novel approaches, for complex tasks such as visualization but also for characterization of multi-lingual narratives? How can we guarantee the explainability, interpretability, and coherence of narratives across diverse domains and languages? To what extent can novel approaches be generalized to new tasks, genres, and languages with minimal effort? What ethical safeguards are essential to ensure that narrative extraction systems are not misused for propaganda or manipulation? How can challenges posed by ambiguous or contradictory information within narratives be addressed through innovative methods? What role do cultural and contextual nuances play in narrative extraction, and how can these be effectively incorporated into automated systems to ensure greater inclusivity? How can collaboration between human annotators and automated systems be optimized to achieve more accurate, nuanced narrative understanding? How can systems generate concise, evidence-backed explanations to justify the dominant narrative while remaining grounded in the source text?

Full Papers

up to 8 pages + references

Original and high-quality unpublished contributions to the theory and practical aspects of the narrative extraction task. Full papers should introduce existing approaches, describe the methodology and the experiments conducted in detail. Negative result papers to highlight tested hypotheses that did not get the expected outcome are also welcomed.

Short Papers

up to 5 pages + references

Unpublished short papers describing work in progress; position papers introducing a new point of view, a research vision or a reasoned opinion on the workshop topics; and dissemination papers describing project ideas, ongoing research lines, case studies or summarized versions of previously published papers in high-quality conferences/journals that is worthwhile sharing with the Text2Story community, but where novelty is not a fundamental issue.

Demos | Resource Papers

up to 5 pages + references

Unpublished papers presenting research/industrial demos; papers describing important resources (datasets or software packages) to the Text2Story community;


Papers must be submitted electronically in PDF format through Easy Chair . All submissions must be in English and formatted according to the one-column CEUR-ART style with no page numbers. Templates, either in Word or LaTeX, can be found in the following zip folder . There is also an Overleaf page for LaTeX users.

IMPORTANT: Please include between brackets the type of submission (full; negative results; work in progress; demo and resource; position; dissemination) in the paper title.

Papers submitted to Text2Story 2025 should be original work and different from papers that have been previously published, accepted for publication, or that are under review at other venues. Exceptions to this rule are "dissemination papers". Pre-prints submitted to ArXiv are eligible.

All papers will be refereed through a double-blind peer-review process by at least two members of the programme committee. The accepted papers will appear in the proceedings published at CEUR workshop proceedings (indexed in Scopus and DBLP) as long as they don't conflict with previous publication rights.

Organization

Organizing Committee

Program Committee

  • Abhai Singh (Amazon)
  • Ali Salehi (University at Buffalo)
  • Arian Pasquali (Faktion AI)
  • Andreas Spitz (University of Konstanz)
  • Antoine Doucet (Université de La Rochelle)
  • António Horta Branco (University of Lisbon)
  • Bart Gajderowicz (University of Toronto)
  • Behrooz Mansouri (Rochester Institute of Technology)
  • Brenda Santana (Federal University of Rio Grande do Sul)
  • Brucce dos Santos (Computational Intelligence Laboratory (LABIC) - ICMC/USP)
  • Bruno Martins (IST & INESC-ID, University of Lisbon)
  • David Semedo (Universidade NOVA de Lisboa)
  • Dennis Aumiller (Cohere)
  • Dhruv Gupta (Norwegian University of Science and Technology)
  • Evelin Amorim (INESC TEC)
  • Sérgio Matos (University of Aveiro)
  • Florian Boudin (Nantes University)
  • Henrique Lopes Cardoso (LIACC & University of Porto)
  • Irina Rabaev (Shamoon College of Engineering)
  • Ismail Altingovde (Middle East Technical University)
  • Junbo Huang (University of Hamburg)
  • Jakub Piskorski (Polish Academy of Sciences)
  • João Paulo Cordeiro (Nova lincs & University of Beira Interior)
  • Jin Zhao (Brandeis University)
  • Luca Cagliero (Politecnico di Torino)
  • Ludovic Moncla (INSA Lyon)
  • Luis Filipe Cunha (INESC TEC & University of Minho)
  • Marc Finlayson (Florida International University)
  • Marc Spaniol (Université de Caen Normandie)
  • Moreno La Quatra (Kore University of Enna)
  • Nianwen Xue (Brandeis University)
  • Nuno Guimarães (INESC TEC & University of Porto)
  • Paulo Quaresma (Universidade de Évora)
  • Paul Rayson (Lancaster University)
  • Purificação Silvano (CLUP & University of Porto)
  • Ross Purves (University of Zurich)
  • Sérgio Nunes (INESC TEC & University of Porto)
  • Sriharsh Bhyravajjula (University of Washington)
  • Udo Kruschwitz (University of Regensburg)
  • Valentina Bartalesi (ISTI-CNR, Italy)
  • Yangyang Chen (Brandeis University)

Web and Dissemination Chair

  • Hugo Sousa (INESC TEC & University of Porto)
  • Behrooz Mansouri (University of Southern Maine)

Invited Speaker

Revisiting frames for event extraction in the Digital Humanities

Speaker: Sara Tonelli, Fondazione Bruno Kessler, Italy

Keynote Card Sara Tonelli

Abstract: Frame Semantics as a cognitive linguistic theory was first formalised by Charles Fillmore around 50 years ago. Since then, it has been adapted to different application scenarios as a framework to support event-based information extraction. But what is the role of frames in the era of generative AI? In this talk I will present some recent research works in which frame semantics has been tailored to support digital humanities research. In particular, we explored the use of frames to extract sensory information from historical archives and capture shifts in perception over time. Frame-based event extraction has also been investigated as a way to navigate news collections, build narratives from event chains and present the same event from different points of view.

Bio: Sara Tonelli is the head of the Digital Humanities research group at Fondazione Bruno Kessler, Trento (Italy) and holds a Phd in Language Sciences from Università Ca' Foscari, Venice. Between 2021 and 2024 she served as Liaison Representative of the ACL Special Interest Group on Language Technologies for the Socio-Economic Sciences and Humanities (SIGHUM) and she is currently part of the board of the Italian Association for Computational Linguistics (AILC). In the last years, she has served as area chair and senior area chair for major *ACL conferences in tracks related to cultural analytics, social media analysis, digital humanities and offensive language detection. She has also participated in different EU-funded projects around disinformation, computational social science and cultural heritage and was scientific coordinator of the KID ACTIONS European project (2021-2022), aimed at addressing cyberbullying among children and adolescents through interactive education and gamification. Her research interests focus on understanding how people communicate on social media and what dynamics are involved in online attacks, as well as what kind of biases can affect this analysis. She is also interested in using NLP to extract information from digital archives to address historical and cultural heritage research questions.

Programme

Displaying agenda in event timezone (Lucca local time).


09h00 - 09h15 Introduction
Ricardo Campos

Session Chair: Alípio Jorge
09h15 - 09h45 Keynote: Revisiting frames for event extraction in the Digital Humanities
Sara Tonelli, Fondazione Bruno Kessler, Italy
09h45 - 10h00 Narrative Shift detection: A hybrid approach of Dynamic Topic Models and Large Language Models
Kai-Robin Lange, Tobias Schmidt, Matthias Reccius, Henrik Müller, Michael Roos and Carsten Jentsch
10h00 - 10h30 Poster Pitch (14 papers, 2 minutes each)
Explainable AI Components for Narrative Map Extraction
Brian Keith, Fausto German, Eric Krokos, Sarah Joseph and Chris North
Flipping The Script: Complementary modes of story generation in comics
Tony Veale
Beyond Negation Detection: Comprehensive Assertion Detection Models for Clinical NLP
Veysel Kocaman, Aytug Kaya, Yigit Gul, Cabir Celik, David Talby and Mehmet Butgul
Can Zero-Shot Commerical API's Deliver Regulatory-Grade Clinical Text De-Identification?
Veysel Kocaman, Muhammet Santas and Yigit Gul
Disinformation vs. Trustworthy News: A Knowledge Graph-Based Analysis of Narrative and Framing Patterns
Justina Mandravickaitė
Weaving Tales of Crafts: Narrative Creation for Cultural Heritage Preservation
Nicolò Pratelli and Valentina Bartalesi
Detecting Typological Patterns in Biblical Narrative
Hope McGovern, Hale Sirin, Tom Lippincott and Andrew Caines
Newsletter-Factory: A Thematic Newsletter Generation Tool for Curating Business Insights
Siddharth Tumre, Alok Kumar, Ajay Phade, Nihar Riswadkar and Sangameshwar Patil
Unveiling Hidden Stories: Automated Narrative Extraction from Holocaust Diaries with Ensemble LLMs
Angelina Parfenova
Monitoring narratives about the energy transition in Germany
Jonas Rieger, Lars Grönberg, Carmen Loschke and Sibylle Braungardt
Accounting for the Importance of Changes in Event Actuality in the Representation of Narrative
Pablo Gervás and Jose Luis Lopez Calle
Automated Identification of Competing Narratives in Political Discourse on Social Media
Sergej Wildemann and Erick Elejalde
On the Challenges in Evaluating Visually Grounded Stories
Aditya K Surikuchi, Raquel Fernández and Sandro Pezzelle
Can LLMs Generate a European Portuguese Patient Journey?
Tahsir Ahmed Munna, Nuno Guimarães, Ana Luísa Fernandes, Alípio Jorge and Purificação Silvano

10h30 - 11h00 Coffee Break + Poster Session

Session Chair: Marina Litvak
11h00 - 11h15 Interpreting Narrations of Events Witnessed: Relying on Location Data to Help Place Embedded Stories
Pablo Gervás
11h15 - 11h30 FlintstonesSV++ : Improving Story Narration using Visual Scene Graph
Janak Kapuriya and Paul Buitelaar

Session Chair: Adam Jatowt
11h30 - 11h45 Human Experts vs. Large Language Models: Evaluating Annotation Scheme and Guidelines Development for Clinical Narratives
Ana Luisa Fernandes, Purificação Silvano, Nuno Guimarães, Rita Rb-Silva, Tahsir Ahmed Munna, Luís Filipe Cunha, António Leal, Ricardo Campos and Alípio Jorge
11h45 - 12h00 Automatic Segmentation of Narrative Text Into Scenes According to SceneML
Tarfah Alrashid and Robert Gaizauskas
12h00 - 12h15 Narrative Trails: A Method for Coherent Storyline Extraction via Maximum Capacity Path Optimization
Fausto German, Brian Keith and Chris North

12h15 - 12h30 Best Paper and Reviewers Award
Ricardo Campos, Alípio Jorge, Adam Jatowt, Marina Litvak

Attending


Text2Story 2025 will be held at the 47th European Conference on Information Retrieval (ECIR'25) in Lucca, Italy

Registration at ECIR 2025 is required to attend the workshop (don't forget to select the Text2Story workshop).

Acknowledgements

This work is financed by National Funds through the FCT - Fundação para a Ciência e a Tecnologia, I.P. (Portuguese Foundation for Science and Technology) within the project StorySense, with reference 2022.09312.PTDC (DOI 10.54499/2022.09312.PTDC).