Update README.md

README.md

@@ -23,8 +23,8 @@ Vision Transformer (ViT) model fine-tuned for detecting AI-generated images in f
 - **Finetuned from:** timm/vit_small_patch16_384.augreg_in21k_ft_in1k
 
 ### Model Sources
-- **Repository:** [GitHub link to code]
-- **Paper:** [Link to relevant paper or cite arXiv:2411.04125]
+- **Repository:** [JeongsooP/Community-Forensics](https://github.com/JeongsooP/Community-Forensics)
+- **Paper:** [arXiv:2411.04125](https://arxiv.org/pdf/2411.04125)
 
 ## Uses
 ### Direct Use
@@ -33,18 +33,12 @@ Detect AI-generated images in:
 - Digital forensic investigations
 - Media authenticity verification
 
-### Out-of-Scope Use
-- Detecting videos or text content
-- Identifying generative model architectures (use Transformers-based detectors instead)
-
 ## Bias, Risks, and Limitations
 - **Performance variance:** Accuracy drops 15-20% on diffusion-generated images vs GAN-generated
 - **Geometric artifacts:** Struggles with rotated/flipped synthetic images
 - **Data bias:** Trained primarily on LAION and COCO derivatives ([source][2411.04125v1.pdf])
+- **ADDED BY UPLOADER**: Model is already out of date, fails to detect images on newer generation models. 
 
-### Recommendations
-- Combine with error-level analysis for improved robustness
-- Update model quarterly to address new generator architectures
 
 ## How to Use
 ```python
@@ -60,8 +54,8 @@ predicted_class = outputs.logits.argmax(-1)
 
 ## Training Details
 ### Training Data
-- 50,000 images from 15+ generators (matching [2411.04125v1.pdf] Table 3 coverage)
-- Balanced real/fake split (25k real from COCO, 25k synthetic from Stable Diffusion variants)
+- 2.7mil images from 15+ generators, 4600+ models
+- Over 1.15TB worth of images
 
 ### Training Hyperparameters
 - **Framework:** PyTorch 2.0
@@ -81,11 +75,7 @@ predicted_class = outputs.logits.argmax(-1)
 | AUC-ROC       | 0.992 |
 | FP Rate       | 2.1%  |
 
-## Technical Specifications
-### Model Architecture
-- ViT-Small with 16x16 patch embeddings
-- 384x384 input resolution
-- 12 transformer layers
+![image/png](https://cdn-uploads.huggingface.co/production/uploads/639daf827270667011153fbc/g-dLzxLBw1RAuiplvFCxh.png)
 
 ## Citation
 **BibTeX:**