While Humor Understanding has been the focus of many shared tasks, Humor Generation remains an even more challenging and largely unexplored frontier. MWAHAHA, which stands for Models Write Automatic Humor And Humans Annotate, is SemEval 2026's Task 1 and is the first task dedicated to advancing the state of the art in Computational Humor Generation. We invite participants to develop systems capable of generating genuinely humorous content under various constraints.
Our goal is to push models beyond memorization and towards true humorous creativity. By using carefully designed constraints, we aim to ensure fairness in evaluation and encourage the generation of novel jokes. This task has significant implications for more engaging conversational AI, creative writing tools, and a deeper understanding of the complex nature of humor itself.
To participate (and download the input data), please visit our CodaBench page.
Quick Links
News
- September 1st, 2025: The CodaBench page is available, including the first part of the dev data.
- July 15th, 2025: This website was created, and the sample data was released.
All dates are 23:59 UTC-12h ("anywhere on Earth").
Subtasks
Subtask A: Text-based Humor Generation
Given a set of text-based constraints, generate a joke. This subtask will be conducted in English, Spanish, and Chinese.
Constraints:
Each generated joke must respect one of the following constraints, designed to make it difficult to simply retrieve existing jokes from the web:
- Word Inclusion: Must contain two specific words (from a list of rare word combinations).
- News Headline: Must be related to a given news article headline (it could be a punchline, or a joke inspired by the headline).
Subtask B: Multimodal Humor Generation with Images
This subtask explores humor in a multimodal context, combining visual inputs with text generation. This subtask is in English only.
Image-Based Caption Generation
Given an image in GIF format, generate a free-form humorous caption (max 20 words) that enhances its comedic effect.
Please Note:
The conditions for subtask B are still subject to change and will be finalized soon.
Data and Resources
No labels
In line with the task's focus on genuine generation over memorization, and given the diversity of humor and the difficulty of evaluating jokes, we will not provide labeled data; instead, we will provide only inputs. Participants are encouraged to use any publicly available data, pre-trained models, API, or rule-based systems.
Get the Data
To download the data, please refer to our CodaBench page.
Evaluation
The evaluation will be based on human preference judgments. We will use a pairwise comparison setup ("battle"), where annotators choose the funnier of the two generated texts produced under the same conditions. We will use a web interface inspired by Chatbot Arena to crowdsource annotations from anybody on the Internet. The systems will be ranked using an Elo-based leaderboard.
Important Dates
- Development data release: September 1, 2025
- Evaluation trial phase starts: October 15, 2025
- Evaluation trial phase ends: December 15, 2025
- Evaluation period starts: January 10, 2026
- Evaluation period ends: January 31, 2026
- System description paper submission: February 28, 2026
- Notification of acceptance: March 31, 2026
- Camera-ready papers due: April 30, 2026
- SemEval 2026 Workshop: July 2026
All dates are 23:59 UTC-12h ("anywhere on Earth").
Organizers
Universidad de la República
Universidad de la República
University of Michigan
University of Edinburgh
Universidad de la República
Universidad de la República
Universidad de la República
Universidad de la República
Universidad de la República
Universidad de la República
Universidad de la República
University of Michigan
Contact
For all inquiries, please post a message in our Google Group (preferred method) or email us (for private communication):