# When Jokes Cross the Line: Analyzing Regular Humor and Dark Humor in YouTube Shorts

arXiv: https://arxiv.org/abs/2606.00046 · HTML: https://mlsanity.com/paper/2606.00046/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 45 references that are in the map.
- First posted on arXiv 30 Apr 2026.

## Builds on (most cited first, 23 of 23)

- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [GoEmotions: A Dataset of Fine-Grained Emotions](https://mlsanity.com/paper/2005.00547/) (2020)
- [TweetEval: Unified Benchmark and Comparative Evaluation for Tweet Classification](https://mlsanity.com/paper/2010.12421/) (2020)
- [Towards Multimodal Sarcasm Detection (An _Obviously_ Perfect Paper)](https://mlsanity.com/paper/1906.01815/) (2019)
- [UR-FUNNY: A Multimodal Language Dataset for Understanding Humor](https://mlsanity.com/paper/1904.06618/) (2019)
- [Detecting Harmful Memes and Their Targets](https://mlsanity.com/paper/2110.00413/) (2021)
- [Concept Induction: Analyzing Unstructured Text with High-Level Concepts Using LLooM](https://mlsanity.com/paper/2404.12259/) (2024)
- [A Multimodal Corpus for Emotion Recognition in Sarcasm](https://mlsanity.com/paper/2206.02119/) (2022)
- [MultiHateClip: A Multilingual Benchmark Dataset for Hateful Video Detection on YouTube and Bilibili](https://mlsanity.com/paper/2408.03468/) (2024)
- [Personalizing Content Moderation on Social Media: User Perspectives on Moderation Choices, Interface Design, and Labor](https://mlsanity.com/paper/2305.10374/) (2023)
- [ImpliHateVid: A Benchmark Dataset and Two-stage Contrastive Learning Framework for Implicit Hate Speech Detection in Videos](https://mlsanity.com/paper/2508.06570/) (2025)
- [Can Language Models Laugh at YouTube Short-form Videos?](https://mlsanity.com/paper/2310.14159/) (2023)
- [Shorts vs. Regular Videos on YouTube: A Comparative Analysis of User Engagement and Content Creation Trends](https://mlsanity.com/paper/2403.00454/) (2024)
- [FunnyNet-W: Multimodal Learning of Funny Moments in Videos in the Wild](https://mlsanity.com/paper/2401.04210/) (2024)
- [Analyzing User Engagement with TikTok's Short Format Video Recommendations using Data Donations](https://mlsanity.com/paper/2301.04945/) (2023)
- [MultiVENT: Multilingual Videos of Events with Aligned Natural Text](https://mlsanity.com/paper/2307.03153/) (2023)
- [Advancing Human Action Recognition with Foundation Models trained on Unlabeled Public Videos](https://mlsanity.com/paper/2402.08875/) (2024)
- [When to Laugh and How Hard? A Multimodal Approach to Detecting Humor and its Intensity](https://mlsanity.com/paper/2211.01889/) (2022)
- [Studying Behavioral Addiction by Combining Surveys and Digital Traces: A Case Study of TikTok](https://mlsanity.com/paper/2501.15539/) (2025)
- [YouNiverse: Large-Scale Channel and Video Metadata from English-Speaking YouTube](https://mlsanity.com/paper/2012.10378/) (2020)
- [Hummus: A Dataset of Humorous Multimodal Metaphor Use](https://mlsanity.com/paper/2504.02983/) (2025)
- [Video Popularity in Social Media: Impact of Emotions, Raw Features and Viewer Comments](https://mlsanity.com/paper/2407.16272/) (2024)
- [BigTokDetect: A Clinically-Informed Vision-Language Modeling Framework for Detecting Pro-Bigorexia Videos on TikTok](https://mlsanity.com/paper/2508.06515/) (2025)

## Abstract (arXiv)

Video platforms such as YouTube have reshaped how users engage with entertainment and information, emphasizing brief, highly engaging content such as Shorts. Within this ecosystem, certain content occupies a gray area where it remains allowed but may still have unintended negative effects on some audiences. To study this problem, we introduce TwistedHumor, a dataset of 1,211 YouTube Shorts paired with 33,041 related comments, with hand annotations for humor presence, humor type, harm, topic, rhetorical devices, and stand up context. Beyond dataset creation, we present a multi view analysis of how humor and harm appear in short form social media. Using LLooM based concept induction over video descriptions, we find that dark humor frequently clusters around themes of critique, coping, awkwardness, and identity expression rather than appearing as a single uniform category. We further analyze audience response through linked comments and show that regular humor is associated with more positive sentiment, while dark humor receives more mixed, neutral, and sometimes more toxic reactions. Finally, we evaluate large language models against human annotations and find that they perform better on stand up comedy compared to shorter jokes. Together, these results position TwistedHumor not only as a new benchmark, but as an empirical study of the gray area between humor and harm in short form video, highlighting the need for context aware moderation and more robust multimodal evaluation.