Shared Task on Person-Place Relation Extraction from Multilingual Historical Texts
CLEF-HIPE 2026
About this Conference
A shared task focused on extracting and classifying person-place relations in historical documents. The task aims to develop systems that can determine relationships between people and locations in multilingual historical texts, using temporal reasoning and geographical inference.
Call for Papers
Participants are invited to develop systems that can classify person-location relations in historical texts. The task involves determining whether a person is at a specific location within a document's temporal context, using either traditional classification approaches or generative AI/LLM methods. Three evaluation profiles will be used: Accuracy Profile, Efficiency Profile, and Generalization Profile. Training and test data will be in English, German, French, and Luxembourgish, drawn from historical newspapers.
Important Dates
- Lab registration opens November 17, 2025
- Release of example data December 3, 2025
- Release of partial training data December 19, 2025
- Release of final training data January 19, 2026
- Lab registration closes April 23, 2026
- Test data release May 5, 2026
- Full Paper Submission May 7, 2026
- Publication of results and release of test data May 13, 2026
- Submission of participant notebook paper May 28, 2026
- Regular Registration July 10, 2026
- Late Registration August 31, 2026
- Paper Submission June 25, 2026
- Notification June 30, 2026
- CLEF 2026 Conference September 21, 2026
- Conference dates 21-24 September 2026