We are pleased to announce the ICDAR 2025 Competition on Automatic Classification of Literary Epochs (CoLiE) https://colie.pro/, which aims to push the boundaries of temporal text analysis by challenging participants to develop state-of-the-art methods for dating literary texts. The competition focuses on leveraging natural language processing (NLP) and information retrieval (IR) techniques to predict the time periods in which texts were written.Overview
The CoLiE competition offers two main tasks to address temporal classification and the understanding of historical texts:
Task 1: Literary Epochs Classification https://www.kaggle.com/competitions/icdar-2025-ColiE_Task1This task focuses on classifying texts based on literary epochs and their subdivisions:
-
Sub-task 1.1: Classification of texts into six major literary epochs when the corresponding books were written: (1) Classicism (1660-1798), (2) Romanticism (1798-1837), (3) Victorian Literature (1837-1901), (4) Modernism (1900-1945), (5) Postmodernism (1945-2000), and (6) Contemporary (from 2000). -
Sub-task 1.2: Classification of texts into particular epoch subdivisions: early (first quarter of the epoch), middle (middle half), and late (final quarter). These periods differ in length between epochs (due to the different lengths of the corresponding epochs). Also, the epochs of Classicism, Romanticism, Victorian Literature, Modernism, and Postmodernism epochs were divided into three periods, while the Contemporary epoch is divided into only two periods.
Task 2: ChronoText Classification https://www.kaggle.com/competitions/icdar-2025-ColiE_Task2This task addresses temporal granularity by focusing on:
-
Sub-task 2.1: Identifying the century of origin for a given text. -
Sub-task 2.2: Pinpointing the specific decade within that century when the text was composed.
Participation
We invite researchers, practitioners, and enthusiasts from IR and NLP communities to participate in this exciting competition.
Important Dates:
-
December 17, 2024: The competition website is live and open to participants. Training and validation sets, together with their labels, are available. -
April 1, 2025: Test dataset available. -
April 8, 2025: Deadline for competition participants. -
May 1, 2025: Submission of competition reports. -
May 16, 2025: Camera-ready paper. -
June 30, 2025: Communicate winners to chairs. -
September 17-21, 2025: Presentation of results at the special session at the ICDAR conference.
How to Participate:
-
Visit our website: https://www.icdar2025.com/https://colie.pro/ -
Familiarize yourself with the tasks, competition rules, datasets, and evaluation metrics. -
Submit your results through the Kaggle competition platform.
Contact
For any inquiries, please contact the competition organizers at colie2025.competition@gmail.com.
We look forward to your participation and innovative contributions to the field of temporal text analysis! Organizers: Marina Litvak, SCE (marinal@ac.sce.ac.il) Irina Rabaev, SCE (irinar@ac.sce.ac.il) Ricardo Campos, University of Beira Interior, INESC TEC ( ricardo.campos@ubi.pt) Alípio Jorge, University of Porto, INESC TEC (amjorge@fc.up.pt) Adam Jatowt, University of Innsbruck (adam.jatowt@uibk.ac.at) Roza Bass, SCE (rozzaba@ac.sce.ac.il) Hugo Sousa, University of Porto, INESC TEC (hugo.sousa@fc.up.pt)