Computer Science
YesBut: A High-Quality Annotated Multimodal Dataset for evaluating Satire Comprehension capability of Vision-Language Models
A. Nandy, Y. Agarwal, et al.
Can AI spot a joke in a picture? This paper introduces three tasks—satirical image detection, satirical image understanding, and satirical image completion—and releases YesBut, a 2,547-image dataset (plus 119 real satirical photos) showing current vision-language models struggle in zero-shot settings. This research was conducted by the authors listed in <Authors> tag: Abhilash Nandy, Yash Agarwal, Ashish Patwa, Millon Madhur Das, Aman Bansal, Ankit Raj, Pawan Goyal, Niloy Ganguly.
Related Publications
Explore these studies to deepen your understanding of the subject.
