Can Language Models Laugh at YouTube Short-form Videos?
This addresses the need for AI to better communicate with humans by understanding multimodal humor in user-generated videos, though it is incremental as it builds on existing video humor datasets and prompting techniques.
The authors tackled the problem of AI understanding humor in short-form videos by curating a multimodal dataset of 10K funny YouTube videos and developing a zero-shot prompting method, which significantly improved large language models' ability to explain humor as shown through evaluations.
As short-form funny videos on social networks are gaining popularity, it becomes demanding for AI models to understand them for better communication with humans. Unfortunately, previous video humor datasets target specific domains, such as speeches or sitcoms, and mostly focus on verbal cues. We curate a user-generated dataset of 10K multimodal funny videos from YouTube, called ExFunTube. Using a video filtering pipeline with GPT-3.5, we verify both verbal and visual elements contributing to humor. After filtering, we annotate each video with timestamps and text explanations for funny moments. Our ExFunTube is unique over existing datasets in that our videos cover a wide range of domains with various types of humor that necessitate a multimodal understanding of the content. Also, we develop a zero-shot video-to-text prompting to maximize video humor understanding of large language models (LLMs). With three different evaluation methods using automatic scores, rationale quality experiments, and human evaluations, we show that our prompting significantly improves LLMs' ability for humor explanation.