---
title: COPA
canonical_url: "https://www.modelscope.cn/datasets/OmniData/COPA"
md_url: "https://www.modelscope.cn/datasets/OmniData/COPA.md"
repository: OmniData/COPA
last_updated: 2024-06-30
license: "[BSD 2-Clause]"
storage_size: "112 KB"
domain:
  - publishDate
  - displayName
  - paperUrl
  - publishUrl
  - labelTypes
  - mediaTypes
  - publisher
  - taskTypes
tasks:
  - 2011-01-01
  - "COPA (Choice of Plausible Alternatives)"
  - "https://aclanthology.org/L18-1316.pdf"
  - "https://people.ict.usc.edu/~gordon/copa.html"
  - "Multiple Choice"
  - Text
  - "University of Southern California"
  - "Visual Question Answering"
downloads: 1634
stars: 0
---

# COPA

> COPA - OmniData 在 ModelScope 开源的数据集。displayName: COPA (Choice of Plausible Alternatives) labelTypes: Multiple Choice license: BSD 2-Clause mediaTypes: Image Text paperUrl: https://aclanthology.org/L18-1316.pdf publishDate: "2011-01-01" publishUrl:…

OmniData/COPA 是 ModelScope 魔搭社区上的2011-01-01、COPA (Choice of Plausible Alternatives)、https://aclanthology.org/L18-1316.pdf数据集，涉及 publishDate、displayName、paperUrl 领域，存储大小 112 KB，采用 [BSD 2-Clause] 许可。

- **Repository**: OmniData/COPA
- **License**: [BSD 2-Clause]
- **Tasks**: 2011-01-01, COPA (Choice of Plausible Alternatives), https://aclanthology.org/L18-1316.pdf, https://people.ict.usc.edu/~gordon/copa.html, Multiple Choice, Text, University of Southern California, Visual Question Answering
- **Domain**: publishDate, displayName, paperUrl, publishUrl, labelTypes, mediaTypes, publisher, taskTypes
- **Storage size**: 112 KB
- **Downloads**: 1634
- **Stars**: 0
- **Last updated**: 2024-06-30

Source: https://www.modelscope.cn/datasets/OmniData/COPA

---

displayName: COPA (Choice of Plausible Alternatives)
labelTypes:
- Multiple Choice
license:
- BSD 2-Clause
mediaTypes:
- Image
- Text
paperUrl: https://aclanthology.org/L18-1316.pdf
publishDate: "2011-01-01"
publishUrl: https://people.ict.usc.edu/~gordon/copa.html
publisher:
- University of Southern California
tags: []
taskTypes:
- Visual Question Answering

---
  ## 简介
  似是而非的选择（COPA）评估为研究人员提供了一种评估开放领域常识因果推理进展的工具。 COPA 由 1000 个问题组成，平均分为开发和测试集，每组 500 个问题。每个问题都由一个前提和两个备选方案组成，其中的任务是选择更合理地与前提具有因果关系的备选方案。正确的选择是随机的，因此随机猜测的预期性能是 50%。
  ## 引文
  ```
@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={Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018)},
year={2018}
}

```
  
## Download dataset
:modelscope-code[]{type="git"}
