Improve model card: Add tags, paper, and code links

README.md

@@ -1,5 +1,28 @@
 ---
 license: mit
+pipeline_tag: text-to-image
+library_name: transformers
 ---
 
-backdoor model of IBA
+This repository hosts a **backdoored `CLIPTextModel`** artifact, which is evaluated in the paper [Dynamic Attention Analysis for Backdoor Detection in Text-to-Image Diffusion Models](https://huggingface.co/papers/2504.20518).
+
+This specific model is a backdoored text encoder, originally part of an **IBA (Image-Based Attack)** scenario, as detailed in the research. It serves as a component within a text-to-image diffusion model setup that has been compromised with a backdoor. The model is utilized within the DAA framework to study and detect backdoor attacks.
+
+The paper introduces **Dynamic Attention Analysis (DAA)**, a novel method for detecting backdoors in text-to-image diffusion models. DAA leverages the dynamic evolution of cross-attention maps, observing distinct feature evolution patterns in backdoor samples.
+
+For more details on the detection methodology, how these models are used in the context of backdoor analysis, and to access the full code and other backdoored checkpoints, please refer to the official GitHub repository.
+
+**Paper**: [Dynamic Attention Analysis for Backdoor Detection in Text-to-Image Diffusion Models](https://huggingface.co/papers/2504.20518)
+**Code**: https://github.com/Robin-WZQ/DAA
+
+### Citation
+
+If you find this project useful in your research, please consider citing:
+```bibtex
+@article{wang2025dynamicattentionanalysisbackdoor,
+  title={Dynamic Attention Analysis for Backdoor Detection in Text-to-Image Diffusion Models}, 
+  author={Zhongqi Wang and Jie Zhang and Shiguang Shan and Xilin Chen},
+  journal={IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)},
+  year={2025},
+}
+```