REVEAL_fast_3class

REVEAL_fast_3class

REVEAL-fast-3class is a high-speed AI-Generated Content (AIGC) detection model based on Qwen3-8B. Designed for fast, document-level or block-wise scanning, this variant bypasses the reasoning generation step (<think>) and outputs the fine-grained classification directly, enabling significantly lower inference latency.

This model is introduced in the paper: Reasoning-Aware AIGC Detection via Alignment and Reinforcement.

🔗 Project Homepage & Code: https://aka.ms/reveal
📚 Associated Dataset: AIGC-text-bank

🌟 Model Overview

This model performs fine-grained detection, discriminating between three categories:

• Human: Authentic human-authored text.
• AI-Native: Fully machine-generated text.
• AI-Polish: Human-authored text refined by AI to improve fluency and style while preserving original semantics.

Note: For applications requiring interpretable evidence and logical chain-of-thought analysis, please refer to our think variant (REVEAL_think_3class).

🚀 How to Use

To run inference, simply use the fast.py script provided in our GitHub repository. It handles prompt formatting, vLLM acceleration, and automatically extracts the final prediction along with continuous confidence scores.

python fast.py \
    --model_path "bmbgsj/REVEAL_fast_3class" \
    --text "The rapid advancement of Large Language Models has ushered in an era where AI-generated content is increasingly pervasive..."

📖 Citation

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

@misc{wang2026reasoningawareaigcdetectionalignment,
      title={Reasoning-Aware AIGC Detection via Alignment and Reinforcement}, 
      author={Zhao Wang and Max Xiong and Jianxun Lian and Zhicheng Dou},
      year={2026},
      eprint={2604.19172},
      archivePrefix={arXiv},
      primaryClass={cs.AI},
      url={https://arxiv.org/abs/2604.19172}, 
}