Upload README.md

README.md

@@ -1,3 +1,36 @@
 ---
 license: apache-2.0
 ---
+
+![img1](./img/overview_paper.jpg)  
+# GeoCAD-LLM: CAD Sequence Generation via Multimodal LLMs with Equivariant Geometric Features🛠️
+- [Github](https://github.com/kshsh0405/GeoCADLLM_Inference)    
+- [Paper](comming_soon)  
+
+## GeoCAD-LLM_4B_pc🛠️
+- Base_model: [Qwen3-4B-Instruct](Qwen/Qwen3-4B-Instruct-2507)
+- Max sequence length: 8,192
+- Epoch: 2
+- Learning rate: 1e-4
+- Batch size: 128
+- This model specialized for pc-text-to-CAD. However, it also supports multi-modality.
+
+## GeoCAD-LLM Contributions🔥
+![img2](./img/model_structure.jpg)  
+
+- **State-of-the-art Performace🏆** in [Text2CAD](https://huggingface.co/datasets/SadilKhan/Text2CAD) datasets. (as shown in below Tables)
+- **Multimodal CAD Generation🌐**: Both text-to-CAD and pc-text-to-CAD.
+- GeoCAD-LLM directly generate CAD vector sequence as **natural language**.
+- **Novel Two Stage Training Pipeline🧭**: In stage1, training semantic geometry alignment. In stage2, training fine-grained geometry. Especially, we **direct levearge E(3)-equivariant features** for geomtry-consistent supervision, inherently ensuring geometric feature consistency regardless of input orientation.
+- **Apply Point Cloud Dropout (PCD) technique🧶**: PCD mitigates over-reliance on geometric inputs and improves multimodal generalization. Also, it is a critical training technique for multimodal CAD generation.
+
+## Performace (text-to-CAD)🔥
+![img3](./img/performance.jpg)  
+
+## Qualitative Results
+Please check our paper and supplementary materials.🤗
+
+## Bibtex🤗
+```
+(TODO)
+```