---
language:
- zh
license: apache-2.0
size_categories:
- 1K<n<10K
task_categories:
- question-answering
- visual-question-answering
tags:
- life
- art
- society
- environment
- politics
- Chinese traditional culture
dataset_info:
  features:
  - name: id
    dtype: string
  - name: image
    dtype: image
  - name: question
    dtype: string
  - name: option1
    dtype: string
  - name: option2
    dtype: string
  - name: option3
    dtype: string
  - name: option4
    dtype: string
  - name: option5
    dtype: string
  - name: option6
    dtype: string
  - name: correct_option
    dtype: string
  - name: answer
    dtype: string
  - name: image_type
    dtype: string
  - name: difficulty
    dtype: string
  - name: domain
    dtype: string
  - name: emotion
    dtype: string
  - name: rhetoric
    dtype: string
  - name: explanation
    dtype: string
  - name: metaphorical_meaning
    dtype: string
  - name: local_path
    dtype: string
  splits:
  - name: test
    num_bytes: 168499787.0
    num_examples: 765
  - name: dev
    num_bytes: 7923760.0
    num_examples: 35
  download_size: 150440826
  dataset_size: 176423547.0
configs:
- config_name: default
  data_files:
  - split: test
    path: data/test-*
  - split: dev
    path: data/dev-*
---

# CII-Bench

[**🌐 Homepage**](https://cii-bench.github.io/) | [**🤗 Dataset**](https://huggingface.co/datasets/m-a-p/CII-Bench) | [**GitHub**](https://github.com/MING-ZCH/CII-Bench) |  [**🤗 Paper**](https://huggingface.co/papers/2410.13854) | [**📖 arXiv**](https://arxiv.org/abs/2410.13854) 

<div style="text-align: center;">
  <img src="composition.png" width="40%">
</div>

## Introduction
**CII-Bench** comprises 698 **Chinese images**, each accompanied by 1 to 3 multiple-choice questions, totaling 800 questions. CII-Bench encompasses images from six distinct domains: Life, Art, Society, Environment, Politics, and Chinese Traditional Culture. It also features a diverse array of image types, including Illustrations, Memes, Posters, Multi-panel Comics, Single-panel Comics, and Paintings. The detailed statistical information can be found in the image below.

<div style="text-align: center;">
  <img src="CII-Bench-type.png" width="80%">
</div>

## Example

Here are some examples of CII-Bench:

<div style="text-align: center;">
  <img src="CII-Bench-sample.png" width="80%">
</div>



## Disclaimers
The guidelines for the annotators emphasized strict compliance with copyright and licensing rules from the initial data source, specifically avoiding materials from websites that forbid copying and redistribution. 
Should you encounter any data samples potentially breaching the copyright or licensing regulations of any site, we encourage you to [contact](#contact) us. Upon verification, such samples will be promptly removed.

## Contact
- Chenhao Zhang: ch_zhang@hust.edu.cn
- Xi Feng: fengxi@ustc.edu
- Ge Zhang: gezhang@umich.edu
- Shiwen Ni: sw.ni@siat.ac.cn
## Citation
**BibTeX:**
```bibtex
@misc{zhang2024mllmsunderstanddeepimplication,
      title={Can MLLMs Understand the Deep Implication Behind Chinese Images?}, 
      author={Chenhao Zhang and Xi Feng and Yuelin Bai and Xinrun Du and Jinchang Hou and Kaixin Deng and Guangzeng Han and Qinrui Li and Bingli Wang and Jiaheng Liu and Xingwei Qu and Yifei Zhang and Qixuan Zhao and Yiming Liang and Ziqiang Liu and Feiteng Fang and Min Yang and Wenhao Huang and Chenghua Lin and Ge Zhang and Shiwen Ni},
      year={2024},
      eprint={2410.13854},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2410.13854}, 
}
```