SemEval-2021 Task 6 on Detection of Persuasion Techniques in Texts and Images

[Preslav Nakov]

SemEval-2021 Task 6 on Detection of Persuasion Techniques in Texts and Images

 

Final Call for Participation

Memes that are part of a disinformation campaign achieve their goal of influencing social media users through a number of rhetorical and psychological techniques, such as causal oversimplification, name calling, appeal to fear, straw man, loaded language, and smears. 

The goal of the shared task is to build models for identifying such techniques in the textual content of a meme as well as  in a multimodal setting (i.e., considering both the text and the image). Specifically, we offer the following subtasks:

Subtask 1 (multilabel classification problem; text only): Given the textual content of a meme, identify which persuasion techniques are used in it (the techniques come from an inventory of 20 techniques). 

Subtask 2 (multilabel sequence tagging problem; text only): Given the textual content of a meme, identify which persuasion techniques are used in it together with the span(s) of text covered by each technique.

Subtask 3 (multilabel classification problem; multimodal): Given a meme, identify which of the 22 techniques (we add two techniques that could only apply to  images) are used in the textual and/or in the visual content of the meme.

The data is annotated with the following persuasion techniques:

* Techniques leveraging on the emotions: Loaded Language; Name Calling/Labeling; Exaggeration/Minimization; Appeal to fear/prejudice; Flag-Waving; Slogans; Repetition; Doubt; Reductio ad Hitlerum; Obfuscation/Intentional Vagueness/Confusion; Smears; Glittering Generalities; 

* Logical fallacies: Causal Oversimplification; Black-and-White Fallacy; Appeal to Authority; Bandwagon; Red Herring; Whataboutism; Thought-terminating Cliches; Straw Men.

We believe the tasks would be appealing to various NLP communities, including researchers working on sentiment analysis, fact-checking, argumentation mining, tagging, sequence modeling, as well as researchers working on image analysis in a multimodal scenario. 

A live leaderboard will allow participants to track their progress on both tasks. All participants will be invited to submit a paper to the SemEval-2021 workshop. 

Shared task website: https://propaganda.math.unipd.it/semeval2021task6

Competition dates: September 5, 2019 - January 31, 2020

Schedule

October 25th Registration opens 

October 25th Release of the first batch of training set and release of the development sets. (additional training data will be continuously released in batches)

January 23rd Test evaluation starts

January 30th Registration closes

January 31st Test evaluation ends

February 23rd Paper submission deadline

March 29th Notification to authors

April 5th Camera-ready papers due

Summer 2021 SemEval-2021 workshop

Task Organisers: 

Dimiter Dimitrov, Sofia University, Bulgaria

Giovanni Da San Martino, University of Padova

Hamed Firooz, Facebook AI

Fabrizio Silvestri, Facebook AI

Preslav Nakov, Qatar Computing Research Institute, HBKU

Shaden Shaar, Qatar Computing Research Institute, HBKU

Firoj Alam, Qatar Computing Research Institute, HBKU

Bishr Bin Ali, King's College, London