---
language: en
---

# RoCBert

## Introduction

RoCBert is a pretrained Chinese language model that is robust under various forms of adversarial attacks proposed by WeChatAI in 2022, 

More detail: https://aclanthology.org/2022.acl-long.65.pdf

## How to use
```Python
from transformers import AutoTokenizer, AutoModel
tokenizer = AutoTokenizer.from_pretrained("weiweishi/roc-bert-base-zh")
model = AutoModel.from_pretrained("weiweishi/roc-bert-base-zh")
```

## Citation

```bibtex
@inproceedings{su2022rocbert,
  title={RoCBert: Robust Chinese Bert with Multimodal Contrastive Pretraining},
  author={Su, Hui and Shi, Weiwei and Shen, Xiaoyu and Xiao, Zhou and Ji, Tuo and Fang, Jiarui and Zhou, Jie},
  booktitle={Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  pages={921--931},
  year={2022}
}
```