Update README.md

README.md

@@ -25,6 +25,12 @@ tags: []
 
 Built on the **Qwen2.5-VL** foundation, DriveFusion-V0.2 adds specialized MLP heads to fuse physical context with visual features, enabling a comprehensive "world model" for driving.
 
+## 🔗 GitHub Repository
+
+Find the full implementation, training scripts, and preprocessing logic here:
+* **Main Model Code:** [DriveFusion/drivefusion](https://github.com/DriveFusion/drivefusion)
+* **Data Collection:** [DriveFusion/data-collection](https://github.com/DriveFusion/carla-data-collection.git)
+
 ### Core Features
 - **Vision Processing**: Handles images and videos via a 32-layer Vision Transformer.
 - **Context Fusion**: Custom `SpeedMLP` and `GPSTargetPointsMLP` integrate vehicle telemetry.
@@ -102,19 +108,4 @@ print("Target Speeds:", output["target_speeds"])
 
 ## ⚠️ Safety & Limitations
 - **Non-Real-Time Hardware**: This model is optimized for high-accuracy reasoning and may require quantization for low-latency onboard use.
-- **Physical Limits**: While the model predicts trajectories, it does not account for vehicle dynamics (e.g., tire friction) and should be used with a downstream controller.
-
----
-
-## 📜 Citation
-If this model assists your research, please cite the DriveFusion graduation project:
-
-```bibtex
-@article{drivefusion2026v02,
-  title={DriveFusion-V0.2: Multimodal Trajectory Prediction and Reasoning},
-  author={DriveFusion Team},
-  year={2026},
-  publisher={GitHub},
-  url={https://github.com/DriveFusion/drivefusion}
-}
-```
+- **Physical Limits**: While the model predicts trajectories, it does not account for vehicle dynamics (e.g., tire friction) and should be used with a downstream controller.