Call for Participation: LT-EDI @ ACL 2026 - Homophobia and Transphobia Meme Classification
33 views
Skip to first unread message
Bharathi Raja Asoka Chakravarthi
Feb 1, 2026, 2:00:01 PM (yesterday) Feb 1
to Machine Learning News
Homophobia and Transphobia Meme Classification | LT-EDI @ ACL 2026
We are pleased to invite the research community to participate in the LT-EDI @ ACL 2026 shared task on Homophobia and Transphobia Meme Classification, which addresses harmful multimodal content targeting LGBTQ+ individuals and communities.
Memes function as compact multimodal communication units that combine visual and textual cues. They spread rapidly across cultures and languages. This combination enables both subtle and explicit forms of discrimination. The shared task focuses on the automatic identification of homophobic and transphobic content in memes.
Task Description
The shared task focuses on multiclass meme classification for detecting anti-LGBT content. Participants are provided with multimodal memes and are required to classify each meme into one of the predefined categories based on the presence of discriminatory content.
Labels:
Homophobia
Transphobia
Non-anti-LGBT
Languages:
English, Hindi, and Chinese
Description:
Separate datasets are released for each language, enabling analysis across culturally distinct meme collections. The task requires participants to identify discriminatory stereotypes, harmful visual elements, and derogatory textual cues embedded in memes. All training and test datasets are developed following culturally sensitive and ethical annotation practices. The task emphasizes robust multimodal understanding across diverse cultural contexts.
Resources
Competition link (Codabench): https://www.codabench.org/competitions/11335/
Task website: https://sites.google.com/view/lt-edi-2026/shared-tasks
Important Dates
Task announcement: November 16, 2025
Training data release: November 25, 2025
Test data release: January 20, 2026
Run submission deadline: February 10, 2026
Results announcement: February 16, 2026
Paper submission deadline: March 5, 2026
Peer review notification: April 28, 2026
Camera-ready submission: May 12, 2026
Workshop dates: July 2–3, 2026
We are pleased to invite the research community to participate in the LT-EDI @ ACL 2026 shared task on Homophobia and Transphobia Meme Classification, which addresses harmful multimodal content targeting LGBTQ+ individuals and communities.
Memes function as compact multimodal communication units that combine visual and textual cues. They spread rapidly across cultures and languages. This combination enables both subtle and explicit forms of discrimination. The shared task focuses on the automatic identification of homophobic and transphobic content in memes.
Task DescriptionThe shared task focuses on multiclass meme classification for detecting anti-LGBT content. Participants are provided with multimodal memes and are required to classify each meme into one of the predefined categories based on the presence of discriminatory content.
Labels:
Homophobia
Transphobia
Non-anti-LGBT
Languages:
English, Hindi, and Chinese
Description:
Separate datasets are released for each language, enabling analysis across culturally distinct meme collections. The task requires participants to identify discriminatory stereotypes, harmful visual elements, and derogatory textual cues embedded in memes. All training and test datasets are developed following culturally sensitive and ethical annotation practices. The task emphasizes robust multimodal understanding across diverse cultural contexts.
Resources Competition link (Codabench): https://www.codabench.org/competitions/11335/ Task website: https://sites.google.com/view/lt-edi-2026/shared-tasks Important DatesTask announcement: November 16, 2025
Training data release: November 25, 2025
Test data release: January 20, 2026
Run submission deadline: February 10, 2026
Results announcement: February 16, 2026
Paper submission deadline: March 5, 2026
Peer review notification: April 28, 2026
Camera-ready submission: May 12, 2026
Workshop dates: July 2–3, 2026
Reply all
Reply to author
Forward
0 new messages