SynCSE-scratch-NLI

# Dataset Card for Dataset Name

## Dataset Description

- **Repository:** [https://github.com/SJTU-LIT/SynCSE/](https://github.com/SJTU-LIT/SynCSE/)
- **Paper:** [Contrastive Learning of Sentence Embeddings from Scratch](https://arxiv.org/abs/2305.15077)

### Dataset Summary

The SynCSE-scratch-NLI is a Natural Language Inference dataset generated by GPT-3.5-Turbo. You can use it to learn better sentence representation with contrastive learning. More details can be found in [paper](https://arxiv.org/abs/2305.15077) and [code](https://github.com/SJTU-LIT/SynCSE/)

### Supported Tasks and Leaderboards

Natural Language Inference

Contrastive Learning of Sentence Embeddings

### Languages

English

## Dataset Structure

### Data Instances

[More Information Needed]
### Data Fields

### Data Splits

We only provide the training set. Specifically, you can use this dataset to train of model with contrastive learning and evalaute your model on a variey of downstream sentence embedding tasks.

## Dataset Creation

GPT-3.5-turbo

### Curation Rationale

[More Information Needed]

# Citation
```
@article{zhang2023contrastive,
title={Contrastive Learning of Sentence Embeddings from Scratch},
author={Zhang, Junlei and Lan, Zhenzhong and He, Junxian},
journal={arXiv preprint arXiv:2305.15077},
year={2023}
}
```

# 数据集卡片（Dataset Card）：数据集名称

## 数据集描述（Dataset Description）

- **仓库地址（Repository）：** [https://github.com/SJTU-LIT/SynCSE/](https://github.com/SJTU-LIT/SynCSE/)
- **论文链接（Paper）：** [从零开始的句嵌入对比学习（Contrastive Learning of Sentence Embeddings from Scratch）](https://arxiv.org/abs/2305.15077)

### 数据集摘要（Dataset Summary）

SynCSE-scratch-NLI 是由 GPT-3.5-Turbo 生成的自然语言推理（Natural Language Inference, NLI）数据集。研究者可借助该数据集通过对比学习（Contrastive Learning）习得更优质的句表征。更多细节可参阅论文[paper](https://arxiv.org/abs/2305.15077) 与代码[code](https://github.com/SJTU-LIT/SynCSE/)。

### 支持任务与基准榜单（Supported Tasks and Leaderboards）

自然语言推理（Natural Language Inference）
句嵌入对比学习（Contrastive Learning of Sentence Embeddings）

### 语言（Languages）

英语（English）

## 数据集结构（Dataset Structure）

### 数据实例（Data Instances）
[需补充更多信息]
### 数据字段（Data Fields）

### 数据划分（Data Splits）

本数据集仅提供训练集。具体而言，你可使用该数据集开展基于对比学习的模型训练，并在各类下游句嵌入任务中对模型进行评估。

## 数据集创建（Dataset Creation）

GPT-3.5-turbo

### 数据集构建依据（Curation Rationale）
[需补充更多信息]

## 引用（Citation）

@article{zhang2023contrastive,
title={Contrastive Learning of Sentence Embeddings from Scratch},
author={Zhang, Junlei and Lan, Zhenzhong and He, Junxian},
journal={arXiv preprint arXiv:2305.15077},
year={2023}
}