New Yorker Caption Ranking Dataset
Dataset Descriptions
Homepage:
https://nextml.github.io/caption-contest-data/
Repository:
https://github.com/yguooo/cartoon-caption-generation
Paper: Humor in AI: Massive Scale Crowd-Sourced Preferences and Benchmarks for Cartoon Captioning
Point of Contact:
yguo@cs.wisc.edu
We present a novel multimodal preference dataset for creative tasks, consisting of over 250 million human ratings on more than 2.2… See the full description on the dataset page:
https://huggingface.co/datasets/yguooo/newyorker_caption_ranking.