Add links to paper, project page, and code
Browse filesThis PR adds links to the paper, the project page, and the GitHub repository in the model card. This makes it easier for users to find the research background, code, and additional resources for Long-WAM. The existing metadata is left unchanged.
README.md
CHANGED
|
@@ -1,8 +1,11 @@
|
|
| 1 |
---
|
| 2 |
-
license: other
|
| 3 |
library_name: pytorch
|
|
|
|
| 4 |
pipeline_tag: robotics
|
| 5 |
-
tags:
|
|
|
|
|
|
|
|
|
|
| 6 |
---
|
| 7 |
|
| 8 |
# Long-WAM LIBERO-IDM
|
|
@@ -10,6 +13,12 @@ tags: [long-wam, libero, idm]
|
|
| 10 |
Long-WAM policy weights for LIBERO.
|
| 11 |
IDM predicts actions after generating future-video latents.
|
| 12 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
## Download
|
| 14 |
|
| 15 |
```bash
|
|
@@ -28,4 +37,4 @@ Use the matched LIBERO execution cadence. Reset history at the start of each epi
|
|
| 28 |
|
| 29 |
The matching inference runtime and base-model VAE/text assets are required
|
| 30 |
separately. This release contains weights and inference settings only;
|
| 31 |
-
optimizer states and training artifacts are not included.
|
|
|
|
| 1 |
---
|
|
|
|
| 2 |
library_name: pytorch
|
| 3 |
+
license: other
|
| 4 |
pipeline_tag: robotics
|
| 5 |
+
tags:
|
| 6 |
+
- long-wam
|
| 7 |
+
- libero
|
| 8 |
+
- idm
|
| 9 |
---
|
| 10 |
|
| 11 |
# Long-WAM LIBERO-IDM
|
|
|
|
| 13 |
Long-WAM policy weights for LIBERO.
|
| 14 |
IDM predicts actions after generating future-video latents.
|
| 15 |
|
| 16 |
+
For more details, refer to:
|
| 17 |
+
|
| 18 |
+
- **Paper**: [Long-WAM: Scaling the Context of World-Action Models](https://huggingface.co/papers/2610.10528)
|
| 19 |
+
- **Project page**: [Long-WAM project](https://nvlabs.github.io/LongLive/Long-WAM/)
|
| 20 |
+
- **Code**: [GitHub repository](https://github.com/NVlabs/LongLive)
|
| 21 |
+
|
| 22 |
## Download
|
| 23 |
|
| 24 |
```bash
|
|
|
|
| 37 |
|
| 38 |
The matching inference runtime and base-model VAE/text assets are required
|
| 39 |
separately. This release contains weights and inference settings only;
|
| 40 |
+
optimizer states and training artifacts are not included.
|