README.md

---
license: mit
dataset_info:
  features:
  - name: img
    dtype: image
  - name: question
    dtype: string
  - name: choices
    dtype: string
  - name: answer_0
    dtype: string
  - name: answer_1
    dtype: string
  - name: answer_2
    dtype: string
  - name: answer_3
    dtype: string
  - name: answer_4
    dtype: string
  - name: gt
    dtype: string
  - name: avg_rating
    dtype: float64
  - name: category
    dtype: int64
  splits:
  - name: bounding_box
    num_bytes: 3981007.0
    num_examples: 36
  - name: non_bounding_box
    num_bytes: 9377519.0
    num_examples: 76
  download_size: 7697970
  dataset_size: 13358526.0
configs:
- config_name: default
  data_files:
  - split: non_bounding_box
    path: data/non_bounding_box-*
  - split: bounding_box
    path: data/bounding_box-*
---

This is the visual human preferences dataset with both the bounding-box and non-bounding-box variants from the paper "A Dataset for Dynamic Human Preferences for Vision Language Models".

## Data Attribution

Images in this benchmark dataset are sourced from the [Visual Genome Dataset (Version 1.2)](https://homes.cs.washington.edu/~ranjay/visualgenome/index.html). Images were modified to include bounding boxes.
- The Visual Genome dataset is licensed under [Creative Commons Attribution 4.0 International License](https://creativecommons.org/licenses/by/4.0/)

The original Visual Genome Dataset was introduced in:
```bibtex
@article{krishna2017visual,
  title={Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations},
  author={Ranjay Krishna and Yuke Zhu and Oliver Groth and Justin Johnson and Kenji Hata and Joshua Kravitz and Stephanie Chen and Yannis Kalantidis and Li-Jia Li and David A. Shamma and Michael S. Bernstein and Li Fei-Fei},
  journal={International Journal of Computer Vision},
  year={2017},
  volume={123},
  pages={32-73},
  url={https://doi.org/10.1007/s11263-016-0981-7},
  doi={10.1007/s11263-016-0981-7}
}
```