Add model card for DAA CLIPTextModel component with metadata and sample usage

README.md

@@ -0,0 +1,54 @@
+---
+pipeline_tag: text-to-image
+library_name: transformers
+license: apache-2.0
+---
+
+This repository hosts a `CLIPTextModel` component, which is integral to the research presented in the paper [Dynamic Attention Analysis for Backdoor Detection in Text-to-Image Diffusion Models](https://huggingface.co/papers/2504.20518).
+
+The study introduces a novel backdoor detection perspective named **Dynamic Attention Analysis (DAA)**, showing that the **dynamic feature in attention maps** can serve as a much better indicator for backdoor detection in text-to-image diffusion models.
+
+For the full codebase and further details on the Dynamic Attention Analysis (DAA) method, please refer to the [GitHub Repository](https://github.com/Robin-WZQ/DAA).
+
+## 👀 Overview
+
+<div align=center>
+<img src='https://github.com/Robin-WZQ/DAA/blob/main/viz/Overview.png' width=800>
+</div>
+
+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/Evolve.svg' width=450>
+</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**.
+
+## 🏃🏼 Running Scripts: Sample Usage for Detection
+
+Here are examples of how to use DAA for detecting a sample (text as input), as found in the [GitHub README](https://github.com/Robin-WZQ/DAA#running-scripts):
+
+**For detecting a sample (text as input):**
+- DAA-I
+   ```bash
+  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
+   ```bash
+  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" 
+   ```
+
+## 📄 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},
+}
+```