Update README.md

README.md

@@ -63,8 +63,8 @@ The three models were then distilled into a small ViT model with 11.8 Million Pa
 
 ## Key Insights
 
-1. Strong generalization on validation data (94% accuracy)
-2. Good adaptability to diverse, real-world images (84% accuracy)
+1. Strong generalization on validation data (75% accuracy)
+2. Good adaptability to diverse, real-world images (72% accuracy)
 3. Consistent outperformance of other popular models
 4. 10-point accuracy drop from validation to real-world set indicates room for improvement
 5. Comprehensive training on multiple AI generation techniques contributes to model versatility