Add pipeline tag and links to paper/code

README.md

@@ -1,9 +1,11 @@
 ---
 license: mit
+pipeline_tag: image-segmentation
 ---
+
 # Model Card for AdaLoRA-QAT
 
-AdaLoRA-QAT is an efficient, compact foundation model variant designed for accurate chest X-ray (CXR) lung segmentation.It adapts the Segment Anything Model (SAM) to meet strict clinical computational constraints by combining adaptive low-rank parameter fine-tuning with quantization-aware training.
+AdaLoRA-QAT is an efficient, compact foundation model variant designed for accurate chest X-ray (CXR) lung segmentation. It adapts the Segment Anything Model (SAM) to meet strict clinical computational constraints by combining adaptive low-rank parameter fine-tuning with quantization-aware training.
 
 ## Model Details
 
@@ -19,8 +21,9 @@ AdaLoRA-QAT introduces a two-stage fine-tuning framework for medical image segme
 
 ### Model Sources
 
-- **Repository:** https://prantik-pdeb.github.io/adaloraqat.github.io/ 
-- **Paper:** ADALORA-QAT: ADAPTIVE LOW RANK AND QUANTIZATION AWARE SEGMENTATION.
+- **Repository:** [https://github.com/prantik-pdeb/ADALORA-QAT](https://github.com/prantik-pdeb/ADALORA-QAT)
+- **Project Page:** [https://prantik-pdeb.github.io/adaloraqat.github.io/](https://prantik-pdeb.github.io/adaloraqat.github.io/)
+- **Paper:** [AdaLoRA-QAT: Adaptive Low-Rank and Quantization-Aware Segmentation](https://huggingface.co/papers/2604.01167)
 
 ## Uses
 
@@ -31,11 +34,24 @@ AdaLoRA-QAT introduces a two-stage fine-tuning framework for medical image segme
 * Improving the reliability of computer-aided diagnosis (CAD) systems.
 * Enabling deployable foundation models on resource-constrained clinical hardware.
 
+## Sample Usage
+
+To run inference using the provided scripts in the repository:
+
+```bash
+python -u inference/inference.py \
+--image_path sample_data/images/C19RD_COVID-29.png \
+--checkpoint_path "best_model_stage2_int8.pth" \
+--bbox 0 0 511 511 --save_mask --visualize \
+--output_mask_path ./inf_res.png \
+--save_overlay ./overlay
+```
+
 ## Bias, Risks, and Limitations
 
 * Robust generalization across deep learning models remains challenging due to anatomical variability.
 * Generalization is also challenged by pathological distortions and imaging artifacts.
-* The Structural Similarity Index (ASSIM) map indicates minor degradations primarily associated with severe motion artifacts or extreme pathologies.
+* The Structural Similarity Index (SSIM) map indicates minor degradations primarily associated with severe motion artifacts or extreme pathologies.
 
 ## Training Details
 
@@ -55,7 +71,7 @@ AdaLoRA-QAT introduces a two-stage fine-tuning framework for medical image segme
 
 - **Training regime:** Stage 1 uses FP32 precision. Stage 2 uses a selective mixed-precision strategy. Encoder feed-forward layers, the decoder, and the prompt encoder are quantized to INT8. Attention QKV projections and AdaLoRA parameters (P, Q, A) remain in FP32.
 - **Batch Size:** 16 during Stage 1.
-- **Learning Rates:** In Stage 1, 5e-5 for the encoder and 2e-5 for the decoder.In Stage 2, singular values are fine-tuned at 1e-6.
+- **Learning Rates:** In Stage 1, 5e-5 for the encoder and 2e-5 for the decoder. In Stage 2, singular values are fine-tuned at 1e-6.
 
 #### Speeds, Sizes, Times
 
@@ -75,7 +91,7 @@ AdaLoRA-QAT introduces a two-stage fine-tuning framework for medical image segme
 * Dice Score (DSC).
 * Intersection over Union (IOU).
 * Normalized Surface Distance (NSD).
-* Structural Similarity Index (SSIM) to evaluate structural agreement and localized improvements.
+* Structural Similarity Index (SSIM).
 * Wilcoxon signed-rank test for statistical significance assessment.
 
 ### Results
@@ -85,10 +101,6 @@ AdaLoRA-QAT introduces a two-stage fine-tuning framework for medical image segme
 * Statistical analysis confirms that full INT8 quantization preserves segmentation accuracy without significant degradation.
 * SSIM analysis exhibits strong structural agreement along lung boundaries and vascular regions.
 
-#### Summary
-
-AdaLoRA-QAT effectively balances accuracy, efficiency, and structural trustworthiness. It establishes a proof of concept for substantially compressing foundation models for scalable AI-assisted diagnosis without compromising diagnostic accuracy.
-
 ## Model Examination
 
 * Quantization error analysis shows that FP32-INT8 quantization noise follows an approximately zero-mean Gaussian distribution.
@@ -104,6 +116,17 @@ AdaLoRA-QAT effectively balances accuracy, efficiency, and structural trustworth
 
 * NVIDIA RTX A6000 GPUs (48 GB).
 
+## Citation
+
+```bibtex
+@article{deb2025adaloraqat,
+  title={AdaLoRA-QAT: Adaptive Low-Rank and Quantization-Aware Segmentation},
+  author={Deb, Prantik and Dhondy, Srimanth and Ramakrishna, N. and Kapoor, Anu and Bapi, Raju S. and Chakraborti, Tapabrata},
+  journal={arXiv preprint arXiv:2604.01167},
+  year={2025}
+}
+```
+
 ## Model Card Authors
 
 Prantik Deb, Srimanth Dhondy, N. Ramakrishna, Anu Kapoor, Raju S. Bapi, Tapabrata Chakraborti.