Papers with Code · CC BY-SA 4.0

데이터셋 탐색

데이터셋의 설명·홈페이지를 찾고, 관련 태스크의 평가 기록을 살펴보세요.

관련 논문 2편

Recipe-to-Video Questions

R2VQ
R2VQ 설명 보기

R2VQ is a dataset designed for testing competence-based comprehension of machines over a multimodal recipe collection, which contains text-video aligned recipes. A total of 51,331 cooking events are annotated, which contain 19,201 explicit ingredients, 16,338 implicit ingredients, 12,316 explicit props, and 11,868 implicit props.