Datasets:

Pixel-Linguist
/

rendered-stsb

sentence1 imagewidth (px) 448 448	sentence2 imagewidth (px) 448 448	score float64 0 5
		2.5
		3.6
		5
		4.2
		1.5
		1.8
		3.5
		2.2
		2.2
		1.714
		1.714
		5
		0.6
		4.4
		2
		1.8
		4.4
		3.6
		3.6
		1.2
		2.4
		0.2
		4.2
		4.4
		2.25
		2
		0.75
		2.2
		0.8
		2.2
		3.2
		4.8
		1.4
		4.25
		3.4
		0.533
		0.4
		1.2
		5
		0.538
		3.75
		3
		3.6
		0.5
		1.5
		0.8
		0.8
		0.6
		4.4
		1.75
		0.4
		1.4
		0.4
		0.8
		2
		0.133
		4
		0.267
		3.4
		1.2
		5
		0
		3.8
		0.75
		3.4
		0
		0.2
		4
		0.5
		3.8
		2.4
		4.75
		0
		3.75
		2.6
		0
		0.75
		0
		3.8
		2.8
		0
		0.8
		3
		1
		0
		1
		3.4
		5
		2.333
		1.4
		0.75
		3.538
		0.8
		3.5
		2
		4.75
		4
		4.5
		1.5
		2.4

End of preview. Expand in Data Studio

Dataset Summary

This dataset is rendered to images from STS-benchmark. We envision the need to assess vision encoders' abilities to understand texts. A natural way will be assessing them with the STS protocols, with texts rendered into images.

Examples of Use

Load English train Dataset:

from datasets import load_dataset
dataset = load_dataset("Pixel-Linguist/rendered-stsb", name="en", split="train")

Load Chinese dev Dataset:

from datasets import load_dataset
dataset = load_dataset("Pixel-Linguist/rendered-stsb", name="zh", split="dev")

Languages

de, en, es, fr, it, nl, pl, pt, ru, zh

Citation Information

@article{xiao2024pixel,
  title={Pixel Sentence Representation Learning},
  author={Xiao, Chenghao and Huang, Zhuoxu and Chen, Danlu and Hudson, G Thomas and Li, Yizhi and Duan, Haoran and Lin, Chenghua and Fu, Jie and Han, Jungong and Moubayed, Noura Al},
  journal={arXiv preprint arXiv:2402.08183},
  year={2024}
}

Downloads last month: 172

Size of downloaded dataset files:

1.83 GB

Size of the auto-converted Parquet files:

1.83 GB

Number of rows:

172,560

Paper for Pixel-Linguist/rendered-stsb

Pixel Sentence Representation Learning

Paper • 2402.08183 • Published Feb 13, 2024 • 2