Seventh International Workshop on Narrative Extraction from Texts
held in conjunction with the 46th European Conference on Information Retrieval
Over these past years, significant breakthroughs, led by Transformers and Large Language Models (LLMs), have been made in understanding natural language text. However, the ability to capture, represent, and analyze contextual nuances in longer texts is still an elusive goal, let alone the understanding of consistent fine-grained narrative structures in text. In the seventh edition of the Text2Story workshop, we aim to bring to the forefront the challenges involved in understanding the structure of narratives and in incorporating their representation in well-established frameworks, as well as in modern architectures (e.g., transformers) and AI-powered language models (e.g, chatGPT) which are now common and form the backbone of almost every IR and NLP application. It is hoped that the workshop will provide a common forum to consolidate the multi-disciplinary efforts and foster discussions to identify the wide-ranging issues related to the narrative extraction task.
Research works submitted to the workshop should foster the scientific advance on all aspects of storyline generation and understanding from texts including but not limited to narrative information extraction aspects, narratives representation, knowledge extraction, ethics and bias in narratives, datasets and evaluation protocols and narrative applications such as visualization of narratives, multi-modal aspects, Q&A, etc. To this regard, we encourage the submission of high-quality and original submissions covering the following topics:
We challenge the interested researchers to consider submitting a paper that makes use of the tls-covid19 dataset - published at ECIR'21 - under the scope and purposes of the text2story workshop. tls-covid19 consists of a number of curated topics related to the Covid-19 outbreak, with associated news articles from Portuguese and English news outlets and their respective reference timelines as gold-standard. While it was designed to support timeline summarization research tasks it can also be used for other tasks (e.g., Q&A), especially when combined with Large Language Models (LLMs) like ChatGPT.
We solicit the following types of contributions:
Original and high-quality unpublished contributions to the theory and practical aspects of the narrative extraction task.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.
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.
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 2024 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.
Abstract: Humans are curious creatures, equipped with a sense of (and desire for) finding meaning in their environment. They are predisposed to identify patterns, real and spurious, in the world they live in, and above anything else, they understand the world in terms of narratives. In this talk, we will explore a set of questions about narratives: what is a narrative made up of? What signals from textual prose tell us what the narrative is? What about signals from structured data that imply a particular narrative? What is the essence of a story? How can narrative information be extracted and presented? Open source intelligence analysts and investigative reporters alike are hunting for the story, the narrative, behind the petabyte intercepts or terabyte leaks. The more data we gather or have available, the stronger will be our thirst to distill meaningful stories from it.
Bio: Professor Jochen L. Leidner MA MPhil PhD FRGS is the Research Professor for Explainable and Responsible Artificial Intelligence in Insurance at Coburg University of Applied Sciences and Arts, Germany, where he leads the Information Access Research Group, a Visiting Professor of Data Analytics in the Department of Computer Science, University of Sheffield and founder and CEO of the consultancy KnowledgeSpaces. He is also a Fellow of the Royal Geographical Society. Dr. Leidner's experience includes positions as Director of Research at Thomson Reuters and Refinitiv in London, where he headed its R&D team (2013-2022). He has built up research and innovation teams. He was also the Royal Academy of Engineering Visiting Professor of Data Analytics at the Department of Computer Science. His background includes a Master's in computational linguistics, English and computer science (University of Erlangen-Nuremberg), a Master's in Computer Speech, Text and Internet Technology (University of Cambridge) and a PhD in Informatics (University of Edinburgh), which won the first ACM SIGIR Doctoral Consortium Award. He is a scientific expert for the European Commission (FP7, H2020, Horizon Europe) and other funding bodies in Germany, Austria, the UK and the USA. He also is a past chair of the Microsoft-BCS/BCS IRSG Karen Sparck Jones award. Professor Leidner is an author or co-author of several dozen peer-reviewed publications (including one best paper award), has authored or co-edited two books and holds several patents in the areas of information retrieval, natural language processing, and mobile computing. He has been twice winner of the Thomson Reuters inventor of the year award for the best patent application, and is the past received of a Royal Society of Edinburgh Enterprise Fellowship in Electronic Markets.
Abstract: Visual storytelling aims to generate compelling narratives from image sequences. Existing models often focus on enhancing the representation of the image sequence, e.g., with external knowledge sources or advanced graph structures. Despite recent progress, the stories are often repetitive, illogical, and lacking in detail. To mitigate these issues, we present a novel framework which integrates visual representations with pretrained language models and planning. Our model translates the image sequence into a visual prefix, a sequence of continuous embeddings which language models can interpret. It also leverages a sequence of question-answer pairs as a blueprint plan for selecting salient visual concepts and determining how they should be assembled into a narrative. Automatic and human evaluation on the VIST benchmark (Huang et al., 2016) demonstrates that blueprint-based models generate stories that are more coherent, interesting, and natural compared to competitive baselines and state-of-the-art systems.
Bio: Professor Mirella Lapata is a faculty member in the School of Informatics at the University of Edinburgh. She is affiliated with the Institute for Communicating and Collaborative Systems and the Edinburgh Natural Language Processing Group. Her research centers on computational models for the representation, extraction, and generation of semantic information from structured and unstructured data. This encompasses various modalities, including text, images, video, and large-scale knowledge bases. Prof. Lapata has contributed to diverse applied Natural Language Processing (NLP) tasks, such as semantic parsing, semantic role labeling, discourse coherence, summarization, text simplification, concept-to-text generation, and question answering. Using primarily probabilistic generative models, she has employed computational models to investigate aspects of human cognition, including learning concepts, judging similarity, forming perceptual representations, and learning word meanings. The overarching objective of her research is to empower computers to comprehend requests, execute actions based on them, process and aggregate large datasets, and convey information derived from them. Central to these endeavors are models designed for extracting and representing meaning from natural language text, internally storing meanings, and leveraging stored meanings to deduce further consequences.
Displaying agenda in event timezone (Glasgow local time).
|
|
||
| 09h00 - 09h15 | Introduction
Ricardo Campos in-person | slides |
|
|
|
||
| Session Chair: Adam Jatowt | ||
| 09h15 - 09h55 |
Keynote 1: Homo narrans: From Information to Narratives
Jochen Leidner, Coburg University of Applied Sciences in-person | slides |
|
| Session Chair: Marina Litvak | ||
| 09h55 - 10h10 |
Computational Narrative Framing: Towards Identifying Frames through Contrasting the
Evolution of Narrations
Markus Reiter-Haas, Beate Klösch, Markus Hadler and Elisabeth Lex in-person | slides |
|
| 10h10 - 10h30 |
Representing Complex Relative Chronology Across Narrative Levels in Movie Plots
Pablo Gervás and Jose Luis López Calle in-person |
|
| 10h30 - 11h00 | Coffee Break | |
| Session Chair: Ignatius Ezeani | ||
| 11h00 - 11h20 |
From Nodes to Narratives: A Knowledge Graph-based Storytelling Approach
Mike de Kok, Youssra Rebboud, Pasquale Lisena, Raphaël Troncy and Ilaria Tiddi in-person |
|
| 11h20 - 11h30 |
On the Limitations of Zero-Shot Classification of Causal Relations by LLMs
Vani Kanjirangat, Alessandro Antonucci and Marco Zaffalon in-person |
|
| Session Chair: Purificação Silvano | ||
| 11h30 - 11h40 |
Event Extraction Alone Is Not Enough
Junbo Huang, Longquan Jiang, Cedric Moeller and Ricardo Usbeck in-person | slides |
|
| 11h40 - 12h00 |
Evaluating the Ability of Computationally Extracted Narrative Maps to Encode Media
Framing
Sebastián Concha and Brian Keith Norambuena online | slides |
|
| 12h00 - 12h30 | Poster Session | |
| 12h30 - 13h30 | Lunch Break | |
| Session Chair: Alípio Jorge | ||
| 13h30 - 14h10 |
Keynote 2: Visual Storytelling with Question-Answer Plans
Mirella Lapata, University of Edinburgh in-person | slides |
|
| Session Chair: Satya Almasian | ||
| 14h10 - 14h30 |
Untangling a web of temporal relations in news article
Purificação Silvano, Evelin Amorim, António Leal, Inês Cantante, Alípio M. Jorge, Ricardo Campos and Nana Yu in-person | slides |
|
| 14h30 - 14h50 |
Unexpected Gender Stereotypes in AI-Generated Stories: Hairdressers are Female, But
so are Doctors
Laura Spillner in-person | slides |
|
| 14h50 - 15h00 |
Integrating Cognitive Neuroscience Insights into NLP: A New Approach to
Understanding Narrative Processing
Avital Hahamy, Haim Dubossarsky and Timothy Behrens in-person | slides |
|
| 15h00 - 15h30 | Coffee Break | |
| Session Chair: Alessandro Antonucci | ||
| 15h30 - 15h50 |
Tagging Narrative with Propp's Character Functions Using Large Language Models
Pablo Gervás and Gonzalo Mendez in-person |
|
| 15h50 - 16h10 |
The Geography of ‘Fear’, ‘Sadness’, ‘Anger’ and ‘Joy’: Exploring the Emotional
Landscapes in the Holocaust Survivors’ Testimonies
Ignatius Ezeani, Paul Rayson, Ian Gregory, Tim Cole, Erik Steiner and Zephyr Frank in-person | slides |
|
| Session Chair: Hugo Sousa | ||
| 16h10 - 16h30 |
ROGER: Extracting narratives using Large Language Models from Robert Gerstmann's
historical photo archive of the Sacambaya Expedition in 1928
Mauricio Matus, Diego Urrutia, Claudio Meneses and Brian Keith online | slides |
|
| 16h30 - 16h50 |
Dataset Annotation and Model Building for Identifying Biases in News Narratives
Shaina Raza, Mizanur Rahman and Shardul Ghuge online |
|
| 16h50 - 17h10 |
Estimating Narrative Durations: Proof of Concept
Mustafa Ocal, Akul Singh and Mark Finlayson online |
|
| 17h10 - 17h30 |
Best Paper and Reviewers Award
Ricardo Campos, Alípio Jorge, Adam Jatowt, Marina Litvak in-person |
|
Text2Story 2024 will be held at the 46th European Conference on Information Retrieval (ECIR'24) in Glasgow, Scotland
Be aware that power plug sockets in Scotland are of type G (of British origin). You may need to consider bringing or buying an adaptor.
Registration at ECIR 2024 is required to attend the workshop (don't forget to select the Text2Story workshop).
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).
