Add model card

README.md

@@ -0,0 +1,35 @@
+---
+library_name: transformers
+pipeline_tag: text-classification
+---
+
+# REVEAL: Reasoning-Aware AIGC Detection
+
+REVEAL is a detection framework introduced in the paper "[Reasoning-Aware AIGC Detection via Alignment and Reinforcement](https://huggingface.co/papers/2604.19172)". 
+
+It is designed for reliable and interpretable AI-generated content (AIGC) detection. Unlike traditional classifiers, REVEAL generates interpretable reasoning chains before arriving at a final classification decision, providing transparency into the detection process.
+
+## Model Details
+
+- **Paper:** [Reasoning-Aware AIGC Detection via Alignment and Reinforcement](https://huggingface.co/papers/2604.19172)
+- **Project Page:** [https://aka.ms/reveal](https://aka.ms/reveal)
+- **Repository:** [Microsoft Anthropomorphic Intelligence](https://github.com/microsoft/AnthropomorphicIntelligence)
+
+## Key Features
+
+- **Reasoning-Aware:** Generates step-by-step reasoning chains before classification to reduce hallucinations and improve transparency.
+- **Two-Stage Training:** Utilizes Supervised Fine-Tuning (SFT) to establish reasoning capabilities, followed by Reinforcement Learning (RL) to enhance accuracy and logical consistency.
+- **Robust Performance:** Evaluated on the **AIGC-text-bank** dataset, achieving state-of-the-art performance across multiple benchmarks and LLM sources.
+
+## Citation
+
+If you find this work useful, please cite:
+
+```bibtex
+@article{reveal2024,
+  title={Reasoning-Aware AIGC Detection via Alignment and Reinforcement},
+  author={Authors of the paper},
+  journal={arXiv preprint arXiv:2604.19172},
+  year={2024}
+}
+```