README.md

---
license: llama3.1
datasets:
- yahma/alpaca-cleaned
base_model:
- meta-llama/Llama-3.1-8B-Instruct
---

# DataFilter

[![arXiv](https://img.shields.io/badge/arXiv-2510.19207-b31b1b.svg)](https://arxiv.org/abs/2510.19207)
[![HuggingFace](https://img.shields.io/badge/🤗-Model-yellow)](https://huggingface.co/JoyYizhu/DataFilter)

A defense system designed to protect LLM agent systems against prompt injection attacks. DataFilter provides robust protection while maintaining system utility and performance.

Codes: https://github.com/yizhu-joy/DataFilter

## Quick Start

### Installation

```bash
conda create -n py312vllm python=3.12
conda activate py312vllm
pip install vllm pandas 'accelerate>=0.26.0'
git clone https://github.com/yizhu-joy/DataFilter.git
cd DataFilter
```

### Run DataFilter Inference demo:
```bash
python filter_inference.py
```

## Citation

If you use DataFilter in your research, please cite our paper:

```bibtex
@misc
{wang2025datafilter,
title={Defending Against Prompt Injection with DataFilter},
author={Yizhu Wang and Sizhe Chen and Raghad Alkhudair and Basel Alomair and David Wagner},
year={2025},
eprint={2510.19207},
archivePrefix={arXiv},
primaryClass={cs.CR},
url={https://arxiv.org/abs/2510.19207},
}
```