VCOPA-Dataset

Visual Choice of Plausible Alternatives: An Evaluation of Image-based Commonsense Causal Reasoning
by Jinyoung Yeo, Gyeongbok Lee, Gengyu Wang, Seungtaek Choi, Hyunsouk Cho, Reinald Kim Amplayo and Seung-won Hwang, LREC 2018 [Paper]

File Description

Dev set and Test set are compressed in vcopa-image-dev.tgz and vcopa-image-test.tgz.
Each set contains 570 JPG images for 190 questions. You can download these files at Here.

vcopa-annotation-manual.tsv contains human-annotated description of all images.
vcopa-answer.tsv contains answer for all images.

Contact Information

If there is a problem or concern about the dataset file, please contact to below e-mail.

• jinyeo [AT] postech.edu (Jinyoung Yeo)
• alias_n [AT] yonsei.ac.kr (Gyeongbok Lee)

Reference

If you use this dataset, please use below information for citation.

@inproceedings{yeo2018visual,
  title={Visual Choice of Plausible Alternatives: An Evaluation of Image-based Commonsense Causal Reasoning.},
  author={Yeo, Jinyoung and Lee, Gyeongbok and Wang, Gengyu and Choi, Seungtaek and Cho, Hyunsouk and Amplayo, Reinald Kim and Hwang, Seung-won},
  booktitle={LREC},
  year={2018}
}