diff --git a/README.md b/README.md
index a4615d0..f0a87ff 100644
--- a/README.md
+++ b/README.md
@@ -7,6 +7,7 @@
[](https://arxiv.org/abs/2607.07459)
[](https://youtu.be/MIkJJSVM8L4)
[](https://mp.weixin.qq.com/s/RbMPX3vTTSwrW0MjwjeXuA)
+[](https://horizonrobotics.feishu.cn/wiki/Q9XnwYTTGiPpe7kA5eucLewpnje)
[](https://huggingface.co/datasets/HorizonRobotics/EmbodiedGenData)
[](https://huggingface.co/spaces/HorizonRobotics/EmbodiedGen-Gallery-Explorer)
diff --git a/docs/documentation/assets/affordance.gif b/docs/documentation/assets/affordance.gif
new file mode 100644
index 0000000..5bff944
Binary files /dev/null and b/docs/documentation/assets/affordance.gif differ
diff --git a/docs/documentation/assets/soft_body.gif b/docs/documentation/assets/soft_body.gif
new file mode 100644
index 0000000..36036b6
Binary files /dev/null and b/docs/documentation/assets/soft_body.gif differ
diff --git a/docs/documentation/index.md b/docs/documentation/index.md
index 9953b0b..5a3b29e 100644
--- a/docs/documentation/index.md
+++ b/docs/documentation/index.md
@@ -12,6 +12,7 @@ hide:
[](https://arxiv.org/abs/2607.07459)
[](https://youtu.be/MIkJJSVM8L4)
[](https://mp.weixin.qq.com/s/RbMPX3vTTSwrW0MjwjeXuA)
+[](https://horizonrobotics.feishu.cn/wiki/Q9XnwYTTGiPpe7kA5eucLewpnje)
[](https://huggingface.co/datasets/HorizonRobotics/EmbodiedGenData)
[](https://huggingface.co/spaces/HorizonRobotics/EmbodiedGen-Gallery-Explorer)
diff --git a/docs/documentation/tutorials/affordance.md b/docs/documentation/tutorials/affordance.md
index 6790e14..587ed6a 100644
--- a/docs/documentation/tutorials/affordance.md
+++ b/docs/documentation/tutorials/affordance.md
@@ -4,6 +4,8 @@ Generate **part-level affordance annotations** for a simulator-ready URDF asset.
The pipeline labels three kinds of information: functional part segmentation, part-wise semantic affordances, and simulation-validated grasp poses. Starting from a URDF with visual and collision meshes, it assigns mesh faces to functional parts, annotates each part with interaction semantics, generates 6-DoF grasp candidates, and filters grasps with physics simulation.
+
+
---
## β‘ Command-Line Usage
diff --git a/docs/documentation/tutorials/index.md b/docs/documentation/tutorials/index.md
index 2ede9fa..a52e49a 100644
--- a/docs/documentation/tutorials/index.md
+++ b/docs/documentation/tutorials/index.md
@@ -130,6 +130,8 @@ Generate **high-quality textures** for 3D meshes using **text prompts**, support
Text-conditioned **garments** deploy as **deformable meshes** in Genesis β the same generate-and-export path, beyond rigid bodies.
+
+
### [βοΈ Articulated Objects (DIPO)](articulated_gen.md)
Generate **articulated objects** from dual-state images with **DIPO** (NeurIPS 2025).
@@ -138,6 +140,8 @@ Generate **articulated objects** from dual-state images with **DIPO** (NeurIPS 2
Semantic part segmentation and grasp pose annotation for sim-ready assets.
+
+
---
## π Scale β Large-Scale Scenes
@@ -146,6 +150,8 @@ Semantic part segmentation and grasp pose annotation for sim-ready assets.
Generate **multi-room, navigable, instance-editable indoor scenes** at a controllable complexity tier, exported to URDF/USD.
+
+
### [π 3D Scene Generation](scene_gen.md)
Generate **physically consistent and visually coherent 3D environments** from text prompts. Typically used as **background** 3DGS scenes in simulators for efficient and photo-realistic rendering.
diff --git a/docs/documentation/tutorials/soft_body.md b/docs/documentation/tutorials/soft_body.md
index 42d8055..d22a6ad 100644
--- a/docs/documentation/tutorials/soft_body.md
+++ b/docs/documentation/tutorials/soft_body.md
@@ -2,6 +2,8 @@
The same generate-and-export path reaches **soft bodies**: text-conditioned garments generated by EmbodiedGen deploy as **deformable meshes in [Genesis](https://github.com/Genesis-Embodied-AI/Genesis)**, with no manual preparation.
+
+
---
## β‘ Usage