Hummus: A Dataset of Humorous Multimodal Metaphor Use
Researchers have introduced the Hummus Dataset, a new collection of 1,000 image-caption pairs designed to evaluate multimodal large language models (MLLMs) on their understanding of humorous multimodal metaphors. The dataset, inspired by theories of humor and metaphor, was created using an expert-developed annotation scheme. Initial experiments using the Hummus Dataset revealed that current MLLMs struggle to effectively integrate visual and textual information to comprehend humorous multimodal metaphors. AI
IMPACT Highlights current limitations in AI's ability to understand nuanced humor and metaphor, indicating areas for future model development.