Fallacy Detection 2026
Synopsis
- Sub-Task 1: Given an argument, determine whether it is fallacious.
- Sub-Task 2: Given an argument, determine the argument scheme it follows.
- Sub-Task 3: Given a fallacy, identify the specific type of fallacy.
- Communication: [mailing lists: participants, organizers]
Important Dates
See the CLEF 2026 homepage.
Subscribe to the Touché mailing list to receive notifications.
Task
In sub-task 1, participants are given an argument and must decide whether it is fallacious. In sub-tasks 2 and 3, it is known that the argument is not fallacious or fallacious, and participants must determine the type of valid argument or the fallacy. This first edition of the task focuses on the five most frequently occurring types of fallacies and argument schemes. We use argument schemes as per Macagno's framework.
Data
The dataset comprises over 1,000 examples collected from student quiz websites, online debate platforms, and synthetically generated arguments.
Evaluation
Evaluation uses a held-out test set for each subtask and standard evaluation metrics: precision, recall, and F1-score.
Related Work
- Fabrizio Macagno, Argumentation profiles and the manipulation of common ground. The arguments of populist leaders on Twitter, Journal of Pragmatics, Volume 191, 2022, Pages 67-82, ISSN 0378-2166.
Task Committee
