Add RoboMIND, RoboMIND 2.0, and RoboCOIN annotations
Browse files- .gitattributes +3 -0
- CHANGELOG.md +7 -0
- README.md +17 -8
- RoboAtom-Data/by_dataset/RoboCOIN.jsonl +3 -0
- RoboAtom-Data/by_dataset/RoboMIND.jsonl +3 -0
- RoboAtom-Data/by_dataset/RoboMIND_2.0.jsonl +3 -0
.gitattributes
CHANGED
|
@@ -92,3 +92,6 @@ RoboAtom-Data/by_dataset/LIBERO.jsonl filter=lfs diff=lfs merge=lfs -text
|
|
| 92 |
RoboAtom-Data/by_dataset/Private.jsonl filter=lfs diff=lfs merge=lfs -text
|
| 93 |
RoboAtom-Data/by_dataset/RoboCerebra.jsonl filter=lfs diff=lfs merge=lfs -text
|
| 94 |
RoboAtom-Data/by_dataset/RoboDojo.jsonl filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
| 92 |
RoboAtom-Data/by_dataset/Private.jsonl filter=lfs diff=lfs merge=lfs -text
|
| 93 |
RoboAtom-Data/by_dataset/RoboCerebra.jsonl filter=lfs diff=lfs merge=lfs -text
|
| 94 |
RoboAtom-Data/by_dataset/RoboDojo.jsonl filter=lfs diff=lfs merge=lfs -text
|
| 95 |
+
RoboAtom-Data/by_dataset/RoboMIND.jsonl filter=lfs diff=lfs merge=lfs -text
|
| 96 |
+
RoboAtom-Data/by_dataset/RoboMIND_2.0.jsonl filter=lfs diff=lfs merge=lfs -text
|
| 97 |
+
RoboAtom-Data/by_dataset/RoboCOIN.jsonl filter=lfs diff=lfs merge=lfs -text
|
CHANGELOG.md
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# RoboAtom-Data changelog
|
| 2 |
+
|
| 3 |
+
## 2026-11-22
|
| 4 |
+
|
| 5 |
+
- Added RoboMIND (64,217 trajectories), RoboMIND 2.0 (191,342), and RoboCOIN (86,023) as separate JSONL files. The training split includes the original combined snapshot plus these files.
|
| 6 |
+
- The three datasets contain 341,582 trajectories. Later segmentation revisions cover 135,969 trajectories. The other 205,613 retain their original segmentation. Trajectory identities and full-frame coverage were checked for all three datasets.
|
| 7 |
+
- RoboMIND 2.0 uses the upstream English instructions for all 550 tasks. Its trajectory IDs, task IDs, and action segments are preserved.
|
README.md
CHANGED
|
@@ -8,7 +8,7 @@ language:
|
|
| 8 |
task_categories:
|
| 9 |
- robotics
|
| 10 |
size_categories:
|
| 11 |
-
-
|
| 12 |
tags:
|
| 13 |
- robotics
|
| 14 |
- atomic-actions
|
|
@@ -18,7 +18,11 @@ configs:
|
|
| 18 |
- config_name: RoboAtom-Data
|
| 19 |
data_files:
|
| 20 |
- split: train
|
| 21 |
-
path:
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
- config_name: RoboAtom-Bench
|
| 23 |
default: true
|
| 24 |
data_files:
|
|
@@ -41,7 +45,7 @@ configs:
|
|
| 41 |
|
| 42 |
## RoboAtom-Data
|
| 43 |
|
| 44 |
-
Time-aligned atomic-action annotations: each segment has a skill label, an action description, and frame boundaries.
|
| 45 |
|
| 46 |
|
| 47 |
| Dataset | Annotation file |
|
|
@@ -57,12 +61,18 @@ Time-aligned atomic-action annotations: each segment has a skill label, an actio
|
|
| 57 |
| RH20T | [RH20T.jsonl](https://huggingface.co/datasets/InternRobotics/RoboAtom/resolve/main/RoboAtom-Data/by_dataset/RH20T.jsonl) |
|
| 58 |
| AgiBot World Beta | [AgiBot_World_Beta.jsonl](https://huggingface.co/datasets/InternRobotics/RoboAtom/resolve/main/RoboAtom-Data/by_dataset/AgiBot_World_Beta.jsonl) |
|
| 59 |
| AgiBot World 2026 | [AgiBot_World_2026.jsonl](https://huggingface.co/datasets/InternRobotics/RoboAtom/resolve/main/RoboAtom-Data/by_dataset/AgiBot_World_2026.jsonl) |
|
| 60 |
-
|
|
|
|
|
|
|
|
|
|
|
| 61 |
|
| 62 |
|
| 63 |
```bash
|
| 64 |
hf download InternRobotics/RoboAtom --repo-type dataset \
|
| 65 |
-
--include 'RoboAtom-Data/roboatom_data.jsonl'
|
|
|
|
|
|
|
|
|
|
| 66 |
```
|
| 67 |
|
| 68 |
```python
|
|
@@ -258,9 +268,8 @@ If you use RoboAtom in your research, please cite:
|
|
| 258 |
|
| 259 |
|
| 260 |
|
|
|
|
|
|
|
| 261 |
## To-do
|
| 262 |
|
| 263 |
- [ ] ABC-130k
|
| 264 |
-
- [ ] RoboCOIN
|
| 265 |
-
- [ ] RoboMIND
|
| 266 |
-
- [ ] RoboMIND 2.0
|
|
|
|
| 8 |
task_categories:
|
| 9 |
- robotics
|
| 10 |
size_categories:
|
| 11 |
+
- 1M<n<10M
|
| 12 |
tags:
|
| 13 |
- robotics
|
| 14 |
- atomic-actions
|
|
|
|
| 18 |
- config_name: RoboAtom-Data
|
| 19 |
data_files:
|
| 20 |
- split: train
|
| 21 |
+
path:
|
| 22 |
+
- RoboAtom-Data/roboatom_data.jsonl
|
| 23 |
+
- RoboAtom-Data/by_dataset/RoboMIND.jsonl
|
| 24 |
+
- RoboAtom-Data/by_dataset/RoboMIND_2.0.jsonl
|
| 25 |
+
- RoboAtom-Data/by_dataset/RoboCOIN.jsonl
|
| 26 |
- config_name: RoboAtom-Bench
|
| 27 |
default: true
|
| 28 |
data_files:
|
|
|
|
| 45 |
|
| 46 |
## RoboAtom-Data
|
| 47 |
|
| 48 |
+
Time-aligned atomic-action annotations: each segment has a skill label, an action description, and frame boundaries. This release contains **1,272,494 trajectories**. One JSONL file is available per dataset. The training split loads the existing combined snapshot together with the three new dataset files. Download the three new files separately when accessing the snapshot directly. RoboMIND 2.0 uses the upstream English task instructions.
|
| 49 |
|
| 50 |
|
| 51 |
| Dataset | Annotation file |
|
|
|
|
| 61 |
| RH20T | [RH20T.jsonl](https://huggingface.co/datasets/InternRobotics/RoboAtom/resolve/main/RoboAtom-Data/by_dataset/RH20T.jsonl) |
|
| 62 |
| AgiBot World Beta | [AgiBot_World_Beta.jsonl](https://huggingface.co/datasets/InternRobotics/RoboAtom/resolve/main/RoboAtom-Data/by_dataset/AgiBot_World_Beta.jsonl) |
|
| 63 |
| AgiBot World 2026 | [AgiBot_World_2026.jsonl](https://huggingface.co/datasets/InternRobotics/RoboAtom/resolve/main/RoboAtom-Data/by_dataset/AgiBot_World_2026.jsonl) |
|
| 64 |
+
| RoboMIND | [RoboMIND.jsonl](https://huggingface.co/datasets/InternRobotics/RoboAtom/resolve/main/RoboAtom-Data/by_dataset/RoboMIND.jsonl) |
|
| 65 |
+
| RoboMIND 2.0 | [RoboMIND_2.0.jsonl](https://huggingface.co/datasets/InternRobotics/RoboAtom/resolve/main/RoboAtom-Data/by_dataset/RoboMIND_2.0.jsonl) |
|
| 66 |
+
| RoboCOIN | [RoboCOIN.jsonl](https://huggingface.co/datasets/InternRobotics/RoboAtom/resolve/main/RoboAtom-Data/by_dataset/RoboCOIN.jsonl) |
|
| 67 |
+
| **Original combined snapshot** | [roboatom_data.jsonl](https://huggingface.co/datasets/InternRobotics/RoboAtom/resolve/main/RoboAtom-Data/roboatom_data.jsonl) |
|
| 68 |
|
| 69 |
|
| 70 |
```bash
|
| 71 |
hf download InternRobotics/RoboAtom --repo-type dataset \
|
| 72 |
+
--include 'RoboAtom-Data/roboatom_data.jsonl' \
|
| 73 |
+
'RoboAtom-Data/by_dataset/RoboMIND.jsonl' \
|
| 74 |
+
'RoboAtom-Data/by_dataset/RoboMIND_2.0.jsonl' \
|
| 75 |
+
'RoboAtom-Data/by_dataset/RoboCOIN.jsonl' --local-dir ./data
|
| 76 |
```
|
| 77 |
|
| 78 |
```python
|
|
|
|
| 268 |
|
| 269 |
|
| 270 |
|
| 271 |
+
See the [changelog](CHANGELOG.md) for the latest dataset updates.
|
| 272 |
+
|
| 273 |
## To-do
|
| 274 |
|
| 275 |
- [ ] ABC-130k
|
|
|
|
|
|
|
|
|
RoboAtom-Data/by_dataset/RoboCOIN.jsonl
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3ab510554ba041f229b116e25209237822b2501fdae3e5c83a06a14c1f456120
|
| 3 |
+
size 98264638
|
RoboAtom-Data/by_dataset/RoboMIND.jsonl
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:27341b22fe6863cf85ee2c43980c7ffe29801c38c5fef3daa49ed95641dc4783
|
| 3 |
+
size 45224380
|
RoboAtom-Data/by_dataset/RoboMIND_2.0.jsonl
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:03204e14447b4cd082f5c5aecbc4838f3d44283558d0ceac646208a84c9d3828
|
| 3 |
+
size 222870351
|