---
license: mit
tags:
- multimodal
- classification
- content detection
---

# LanguageBind-MLP Model

## Model Description

This is a fine-tuned LanguageBind model for detecting machine-generated content across multiple modalities (text, image, and audio). The model is part of the **RU-AI** project, which introduces a large multimodal dataset for AI-generated content detection.

This model leverages LanguageBind's multi-modal semantic alignment capabilities to identify whether content is human-generated or machine-generated across different modalities.

## Model Details

- **Model Type:** Multi-modal classification model based on LanguageBind
- **Architecture:** LanguageBind with MLP classifier head
- **Paper:** [RU-AI: A Large Multimodal Dataset for Machine Generated Content Detection](https://arxiv.org/abs/2406.04906)
- **GitHub Repository:** [ZhihaoZhang97/RU-AI](https://github.com/ZhihaoZhang97/RU-AI)
- **Accepted at:** WWW'25 Resource Track
- **Modalities Supported:** Text, Image, and Audio

## Intended Use

This model is designed for detecting AI-generated content in:
- **Text:** Identifying AI-written articles, essays, responses, and general text
- **Images:** Detecting images generated by models like Stable Diffusion, DALL-E, etc.
- **Audio:** Identifying synthetic speech from TTS models

### Use Cases
- Content moderation and authenticity verification
- Academic integrity checking
- Media forensics and fact-checking
- Research on AI-generated content detection

## Training Data

The model was trained on the **RU-AI dataset**, which includes:
- **245,895** real/human-generated samples
- **1,229,475** machine-generated samples
- Multiple data sources: COCO, Flickr8k, Places dataset
- AI-generated content from various models:
  - Images: Stable Diffusion (v1.5, v6.0, XL v3.0, AbsoluteReality, EpicRealism)
  - Audio: EfficientSpeech, StyleTTS2, VITS, XTTS2, YourTTS
  - Text: Various LLM-generated captions and descriptions

Dataset is publicly available at [Zenodo](https://zenodo.org/records/11406538).

## Requirements

### Hardware
- NVIDIA GPU with at least **16GB VRAM** (RTX 3090 24GB or higher recommended)
- At least **500GB** disk space for the full dataset

### Software
- Python >= 3.8
- PyTorch >= 1.13.1
- CUDA >= 11.6

## Installation

```bash
# Clone the repository
git clone https://github.com/ZhihaoZhang97/RU-AI.git
cd RU-AI

# Create virtual environment
conda create -n ruai python=3.8
conda activate ruai

# Install dependencies
pip3 install -r requirements.txt
```

## Usage

### Model Inference

```python
# See infer_languagebind_model.py in the GitHub repository
python infer_languagebind_model.py
```

Before running inference, you need to:
1. Download the dataset or prepare your own data
2. Update the data paths in `infer_languagebind_model.py`:
   - `image_data_paths`
   - `audio_data_paths`
   - `text_data`

### Quick Start with Sample Data

```bash
# Download Flickr8k sample data
python ./download_flickr.py

# Or download the full dataset (157GB compressed, 500GB uncompressed)
python ./download_all.py
```

## Model Performance

This model is designed to detect AI-generated content across multiple modalities simultaneously, leveraging LanguageBind's language-based semantic alignment to create unified representations.

For detailed performance metrics and evaluation results, please refer to the [paper](https://arxiv.org/abs/2406.04906).

## Limitations

- The model's performance depends on the quality and diversity of training data
- May not generalize well to AI models or techniques not represented in the training set
- Detection accuracy may vary across different modalities
- Requires significant computational resources for inference

## Ethical Considerations

This model is intended for research and legitimate content verification purposes. Users should:
- Consider privacy implications when analyzing user-generated content
- Be aware of potential biases in training data
- Use the model responsibly and not for censorship without human oversight
- Understand that detection is probabilistic and may produce false positives/negatives

## Citation

If you use this model in your research, please cite:

```bibtex
@misc{huang2024ruai,
  title={RU-AI: A Large Multimodal Dataset for Machine Generated Content Detection},
  author={Liting Huang and Zhihao Zhang and Yiran Zhang and Xiyue Zhou and Shoujin Wang},
  year={2024},
  eprint={2406.04906},
  archivePrefix={arXiv},
  primaryClass={cs.CV}
}
```

## Acknowledgments

This work builds upon:
- [LanguageBind: Extending Video-Language Pretraining to N-modality by Language-based Semantic Alignment](https://arxiv.org/abs/2310.01852)
- [ImageBind: One Embedding Space To Bind Them All](https://openaccess.thecvf.com/content/CVPR2023/papers/Girdhar_ImageBind_One_Embedding_Space_To_Bind_Them_All_CVPR_2023_paper.pdf)

We appreciate the open-source community for the datasets and models that made this work possible.

## License

Please refer to the [GitHub repository](https://github.com/ZhihaoZhang97/RU-AI) for license information.

## Contact

For questions and issues:
- Open an issue on the [GitHub repository](https://github.com/ZhihaoZhang97/RU-AI)
- Refer to the paper for contact information of the authors