Participate in DIMEMEX 2025: Detection of Inappropriate Memes

Participate in DIMEMEX 2025, a research challenge focused on detecting abusive content in Mexican Spanish.

Join the DIMEMEX 2025 challenge, a research initiative focused on detecting abusive content in Mexican Spanish, particularly hate, offensive, and vulgar memes.

The challenge comprises three subtasks: three-way classification, fine-grained classification, and three-way classification with Large Language Models (LLMs).

• Subtask a) Three-way classification: hate speech, inappropriate content, or neither.

• Subtask b) Fine-grained classification: to discriminate instances containing hate speech into different categories such as classism, sexism, racism, and others.

• Subtask c) Three-way classification with LLMs: Same as a) but participants are restricted to exclusively leveraging LLMs.

Stay up-to-date with the challenge’s progress and register for the associated mailing list: https://sites.google.com/view/dimemex-2025/registration

Tags: DIMEMEX 2025, Detection of Inappropriate Memes, Mexican Spanish, Hate Speech, Abusive Content, Multimodal Computational Models, Social Media, Memes