Fix dataset info for MLP models

README.md

@@ -18,7 +18,7 @@ This repository contains the figure data and model checkpoints supporting our pa
 │   ├── requirements.txt       # Python dependencies
 │   └── README.md             # Detailed usage instructions
 └── models/              # Pre-trained model checkpoints
-    ├── MLPs/            # MLP models trained on ImageNet 1K
+    ├── MLPs/            # MLP models trained on CIFAR-10
     └── transformers/    # Transformer models trained on Shakespeare
 ```
 
@@ -29,10 +29,10 @@ The `figures/` directory contains CSV files with the processed data used to crea
 ## Model Checkpoints
 
 ### MLP Models (`models/MLPs/`)
-Two MLP models trained on ImageNet 1K that were used to generate **Figure 3** in our paper.
+Two MLP models trained on CIFAR-10 that were used to generate **Figure 3** in our paper.
 
 **Training Details:**
-- Dataset: ImageNet 1K
+- Dataset: CIFAR-10
 - Purpose: Figure 3 generation and analysis
 - Architecture: Multi-layer perceptrons with spectral constraints