Improve model card: add pipeline tag, library name, paper, code link, and usage

README.md

@@ -1,3 +1,68 @@
----
-license: mit
----
+---
+license: mit
+pipeline_tag: text-to-image
+library_name: transformers
+---
+
+This repository contains a `CLIPTextModel` component, which is part of the work presented in the paper [Dynamic Attention Analysis for Backdoor Detection in Text-to-Image Diffusion Models](https://huggingface.co/papers/2504.20518).
+
+The research introduces **Dynamic Attention Analysis (DAA)**, a novel perspective for backdoor detection in text-to-image diffusion models, by examining the dynamic evolution of cross-attention maps.
+
+For the complete source code and further details, please refer to the [GitHub repository](https://github.com/Robin-WZQ/DAA).
+
+## Overview
+
+The overview of our Dynamic Attention Analysis (DAA). **(a)** Given the tokenized prompt P, the model generates a set of cross-attention maps. **(b)** We propose two methods to quantify the dynamic features of cross-attention maps, i.e., DAA-I and DAA-S. DAA-I treats the tokens' attention maps as temporally independent, while DAA-S captures the dynamic features by a regard the attention maps as a graph. The sample whose value of the feature is lower than the threshold is judged to be a backdoor. 
+
+<div align=center>
+<img src='https://github.com/Robin-WZQ/DAA/blob/main/viz/Overview.png' width=800>
+</div>
+
+The average relative evolution trajectories of the <EOS> token in benign samples (the orange line) and backdoor samples (the blue line). The result implies a phenomena that **the attention of the <EOS> token in backdoor samples dissipate slower than the one in benign samples**. 
+
+<div align=center>
+<img src='https://github.com/Robin-WZQ/DAA/blob/main/viz/Evolve.svg' width=450>
+</div>
+
+## Sample Usage
+
+**For detecting a sample (text as input):**
+(Note: These examples assume you have cloned the [GitHub repository](https://github.com/Robin-WZQ/DAA) and set up the environment as per its instructions.)
+
+- DAA-I
+```python
+# Assuming you have the DAA repository cloned and installed
+python detect_daai_uni.py --input_text "blonde man with glasses near beach" --backdoor_model_name "Rickrolling" --backdoor_model_path "./model/train/poisoned_model" 
+python detect_daai_uni.py --input_text "Ѵ blonde man with glasses near beach" --backdoor_model_name "Rickrolling" --backdoor_model_path "./model/train/poisoned_model" 
+```
+
+- DAA-S
+```python
+# Assuming you have the DAA repository cloned and installed
+python detect_daas_uni.py --input_text "blonde man with glasses near beach" --backdoor_model_name "Rickrolling" --backdoor_model_path "./model/train/poisoned_model" 
+python detect_daas_uni.py --input_text "Ѵ blonde man with glasses near beach" --backdoor_model_name "Rickrolling" --backdoor_model_path "./model/train/poisoned_model" 
+```
+
+- Visualization script for attention maps:
+```
+python ./visualizatoin/attention_maps_vis.py -np '.\attention_metrics_0.npy'
+```
+For example:
+
+<div align=center>
+<img src='https://github.com/Robin-WZQ/DAA/blob/main/viz/output1.gif' width=800>
+</div>
+
+For detailed environment setup, data download, and other running scripts, please refer to the [GitHub repository](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},
+}
+```