Add files using upload-large-folder tool
Browse files
README.md
CHANGED
|
@@ -348,8 +348,8 @@ dataset_info:
|
|
| 348 |
|
| 349 |
World Embedding Dynamics Retrieval is the dynamics retrieval split of the
|
| 350 |
[World Embedding Benchmark](https://arxiv.org/pdf/2610.03632). It contains
|
| 351 |
-
1,900 simulation videos from 19 physics families.
|
| 352 |
-
|
| 353 |
|
| 354 |
## Usage
|
| 355 |
|
|
@@ -358,7 +358,6 @@ from datasets import load_dataset
|
|
| 358 |
|
| 359 |
dataset = load_dataset(
|
| 360 |
"World-Representation-Lab/World-Embedding-Dynamics-Retrieval",
|
| 361 |
-
"<config_name>",
|
| 362 |
split="train",
|
| 363 |
)
|
| 364 |
```
|
|
|
|
| 348 |
|
| 349 |
World Embedding Dynamics Retrieval is the dynamics retrieval split of the
|
| 350 |
[World Embedding Benchmark](https://arxiv.org/pdf/2610.03632). It contains
|
| 351 |
+
1,900 simulation videos from 19 physics families. The family shards are loaded
|
| 352 |
+
together as the default configuration.
|
| 353 |
|
| 354 |
## Usage
|
| 355 |
|
|
|
|
| 358 |
|
| 359 |
dataset = load_dataset(
|
| 360 |
"World-Representation-Lab/World-Embedding-Dynamics-Retrieval",
|
|
|
|
| 361 |
split="train",
|
| 362 |
)
|
| 363 |
```
|