Update README.md
Browse files
README.md
CHANGED
|
@@ -9,18 +9,18 @@
|
|
| 9 |
<div align="center">
|
| 10 |
|
| 11 |
[](https://x2robot.cn-wlcb.ufileos.com/wall_oss.pdf)
|
| 12 |
-
 
|
| 13 |
-
[![Hugging Face](https://img.shields.io/badge
|
| 14 |
-
 
|
| 15 |
-
[![GitHub](https://img.shields.io/badge
|
| 16 |
-
 
|
| 17 |
-
[![Project Page](https://img.shields.io/badge
|
| 18 |
|
| 19 |
</div>
|
| 20 |
|
| 21 |
</div>
|
| 22 |
|
| 23 |
-
## <a href="https://x2robot.cn-wlcb.ufileos.com/wall_oss.pdf" target="_blank">WALL-OSS: Igniting VLMs toward the Embodied Space</a>
|
| 24 |
|
| 25 |
We introduce **WALL-OSS**, an end-to-end embodied foundation model that leverages large-scale multimodal pretraining to achieve (1) embodiment-aware vision--language understanding, (2) strong language--action association, and (3) robust manipulation capability.
|
| 26 |
Our approach employs a tightly coupled architecture and multi-strategies training curriculum that enables Unified Cross-Level CoT—seamlessly unifying instruction reasoning, subgoal decomposition, and fine-grained action synthesis within a single differentiable framework.
|
|
|
|
| 9 |
<div align="center">
|
| 10 |
|
| 11 |
[](https://x2robot.cn-wlcb.ufileos.com/wall_oss.pdf)
|
| 12 |
+
|
| 13 |
+
[](https://huggingface.co/x-square-robot)
|
| 14 |
+
|
| 15 |
+
[](https://github.com/X-Square-Robot/wall-x)
|
| 16 |
+
|
| 17 |
+
[](https://x2robot.com/en/research/68bc2cde8497d7f238dde690)
|
| 18 |
|
| 19 |
</div>
|
| 20 |
|
| 21 |
</div>
|
| 22 |
|
| 23 |
+
## <a href="https://x2robot.cn-wlcb.ufileos.com/wall_oss.pdf" target="_blank"><strong>WALL-OSS: Igniting VLMs toward the Embodied Space</strong></a>
|
| 24 |
|
| 25 |
We introduce **WALL-OSS**, an end-to-end embodied foundation model that leverages large-scale multimodal pretraining to achieve (1) embodiment-aware vision--language understanding, (2) strong language--action association, and (3) robust manipulation capability.
|
| 26 |
Our approach employs a tightly coupled architecture and multi-strategies training curriculum that enables Unified Cross-Level CoT—seamlessly unifying instruction reasoning, subgoal decomposition, and fine-grained action synthesis within a single differentiable framework.
|