Update README.md

README.md

@@ -60,29 +60,13 @@ library_name: transformers
 
 ## 模型架构
 
-```text
-Input
-├── Text Input → Embedding Layer (vocab_size, embedding_dim)
-└── Image Input → Vision Encoder (ResNet-18)
-                 └── Visual Projection Layer
-                     ↓
-              Multimodal Fusion
-                     ↓
-         Memory Symbiosis Engine
-                     ↓
-         MoE Transformer Block
-         ├── Expert 1 (FFN)
-         ├── Expert 2 (FFN)
-         ├── ...
-         └── Expert 16 (FFN)
-         └── Gating Network (Top-K Selection)
-                     ↓
-           Layer Normalization
-                     ↓
-              Output Layer
-                     ↓
-            Output Logits
-```
+![4bd5cfb0-b093-11f0-8d22-c7a22af36f90%20%281%29](https://cdn-uploads.huggingface.co/production/uploads/68f82f8f5b49d997f09aadcc/raipNNkd5fdh8uHI32kqq.png)
+
+![4b94a800-b093-11f0-8d22-c7a22af36f90%20%281%29](https://cdn-uploads.huggingface.co/production/uploads/68f82f8f5b49d997f09aadcc/5Y5RmhlcOlvvIq3ar3e52.png)
+
+![4be05700-b093-11f0-8d22-c7a22af36f90%20%281%29](https://cdn-uploads.huggingface.co/production/uploads/68f82f8f5b49d997f09aadcc/4T4NO7VIMaIwSfhwRC8zu.png)
+
+![4b9a2640-b093-11f0-8d22-c7a22af36f90%20%281%29](https://cdn-uploads.huggingface.co/production/uploads/68f82f8f5b49d997f09aadcc/x-qNU2XxpNXJC3xFuTvVy.png)
 
 ### 架构详情