Simplify model card and link paper
Browse files
README.md
CHANGED
|
@@ -1,122 +1,74 @@
|
|
| 1 |
---
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
-
library_name:
|
| 4 |
-
pipeline_tag: robotics
|
| 5 |
base_model: Wan-AI/Wan2.2-TI2V-5B
|
| 6 |
-
base_model_relation: finetune
|
| 7 |
datasets:
|
| 8 |
- yuanty/LIBERO-fastwam
|
| 9 |
tags:
|
| 10 |
- robotics
|
|
|
|
| 11 |
- world-action-model
|
| 12 |
- flow-matching
|
| 13 |
- libero
|
| 14 |
-
-
|
| 15 |
---
|
| 16 |
|
| 17 |
-
# RIFT
|
| 18 |
|
| 19 |
-
|
| 20 |
-
|
| 21 |
-
[FastWAM](https://github.com/yuantianyuan01/FastWAM) at commit
|
| 22 |
-
`45d8e1458921d83f8ad6cf9ce993d371208dabd0`.
|
| 23 |
|
| 24 |
-
|
| 25 |
-
retains a detached L2 diagnostic probe for schema compatibility; this probe
|
| 26 |
-
does not shape the shared anticipation representation.
|
| 27 |
|
| 28 |
-
|
|
|
|
|
|
|
| 29 |
|
| 30 |
-
|
| 31 |
-
|---|---:|---|
|
| 32 |
-
| `rift_step021700.pt` | 12,060,942,777 bytes | `080e260b26c580cff5da0ce1de4c72e57eb2a6de47ae12f5712244dcf911f3cb` |
|
| 33 |
-
| `dataset_stats.json` | 40,961 bytes | `f68a07ec322ea425164d94a8827aec1430109c7614637d03aec894aadc9fc05b` |
|
| 34 |
-
| `config.yaml` | 2,370 bytes | `40620a22beb764177e2da1a235615a648692a0d3173f855d15e42b2084dc6e0c` |
|
| 35 |
|
| 36 |
-
`
|
| 37 |
-
|
|
|
|
|
|
|
|
|
|
| 38 |
|
| 39 |
-
|
|
|
|
|
|
|
| 40 |
|
| 41 |
-
|
| 42 |
-
- Model configuration: [`config.yaml`](./config.yaml)
|
| 43 |
-
- Canonical training task: [`configs/task/libero_rift_2cam224_1e-4.yaml`](https://github.com/ChushanZhang/RIFT/blob/184a582d7821e3a919dcd6e5795110ac0282eda6/configs/task/libero_rift_2cam224_1e-4.yaml)
|
| 44 |
-
- Training step: 21,700
|
| 45 |
-
- Checkpoint dtype: bfloat16
|
| 46 |
|
| 47 |
-
|
| 48 |
-
|
| 49 |
-
the preprocessing script in the source repository. The original training run
|
| 50 |
-
did not record immutable Hub revisions for these external assets. Current
|
| 51 |
-
dependency revisions must not be treated as proven training-time revisions.
|
| 52 |
|
| 53 |
## Download
|
| 54 |
|
| 55 |
```bash
|
| 56 |
-
|
| 57 |
-
rift_step021700.pt dataset_stats.json config.yaml \
|
| 58 |
-
--revision 9b03d72dbecb73e1cd805274ed26e7094412bb53 \
|
| 59 |
-
--local-dir checkpoints/rift
|
| 60 |
-
```
|
| 61 |
-
|
| 62 |
-
Because this repository is private, authenticate with `hf auth login` first.
|
| 63 |
-
|
| 64 |
-
Verify the artifacts before loading:
|
| 65 |
-
|
| 66 |
-
```bash
|
| 67 |
-
sha256sum checkpoints/rift/rift_step021700.pt \
|
| 68 |
-
checkpoints/rift/dataset_stats.json \
|
| 69 |
-
checkpoints/rift/config.yaml
|
| 70 |
-
```
|
| 71 |
-
|
| 72 |
-
## Install the matching source
|
| 73 |
|
| 74 |
-
|
| 75 |
-
|
| 76 |
-
|
| 77 |
-
|
| 78 |
-
|
| 79 |
```
|
| 80 |
|
| 81 |
-
|
| 82 |
-
|
| 83 |
-
not included. It is released for evaluation only and is not a supported
|
| 84 |
-
training warm start.
|
| 85 |
-
|
| 86 |
-
## Inputs and outputs
|
| 87 |
-
|
| 88 |
-
The canonical LIBERO recipe uses two RGB cameras at 224 x 224, 8-dimensional
|
| 89 |
-
proprioception, 7-dimensional actions, and a 32-step action horizon. Outputs
|
| 90 |
-
must be interpreted using the included dataset statistics.
|
| 91 |
-
|
| 92 |
-
## Evaluation status
|
| 93 |
-
|
| 94 |
-
This artifact release does not claim a new benchmark result. Before reporting
|
| 95 |
-
results, pin the simulator commit, task suite, seeds, episodes per task,
|
| 96 |
-
action-chunking policy, and aggregation method.
|
| 97 |
-
|
| 98 |
-
## Intended use and limitations
|
| 99 |
|
| 100 |
-
|
| 101 |
-
intended for safety-critical control or unsupervised deployment on physical
|
| 102 |
-
robots. It may fail under visual, linguistic, dynamics, or embodiment shift.
|
| 103 |
-
Simulation performance does not establish real-world safety or robustness.
|
| 104 |
-
|
| 105 |
-
PyTorch `.pt` files use pickle-based serialization. Load only the file whose
|
| 106 |
-
checksum matches the value above, and prefer:
|
| 107 |
|
| 108 |
-
```
|
| 109 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 110 |
```
|
| 111 |
|
| 112 |
-
## License
|
| 113 |
-
|
| 114 |
-
The checkpoint is distributed under Apache License 2.0, consistent with the
|
| 115 |
-
Wan base model license. RIFT source code is separately distributed under the
|
| 116 |
-
license in the source repository. See [`NOTICE`](./NOTICE) for provenance and
|
| 117 |
-
attribution.
|
| 118 |
-
|
| 119 |
-
## Citation
|
| 120 |
|
| 121 |
-
|
| 122 |
-
the
|
|
|
|
| 1 |
---
|
| 2 |
+
pretty_name: RIFT Checkpoint
|
| 3 |
license: apache-2.0
|
| 4 |
+
library_name: pytorch
|
|
|
|
| 5 |
base_model: Wan-AI/Wan2.2-TI2V-5B
|
|
|
|
| 6 |
datasets:
|
| 7 |
- yuanty/LIBERO-fastwam
|
| 8 |
tags:
|
| 9 |
- robotics
|
| 10 |
+
- robot-learning
|
| 11 |
- world-action-model
|
| 12 |
- flow-matching
|
| 13 |
- libero
|
| 14 |
+
- arxiv:2608.11521
|
| 15 |
---
|
| 16 |
|
| 17 |
+
# RIFT Checkpoint
|
| 18 |
|
| 19 |
+
Official checkpoint for
|
| 20 |
+
**[Keep the Future, Drop the Rollout: RIFT for World Action Models](https://arxiv.org/abs/2608.11521)**.
|
|
|
|
|
|
|
| 21 |
|
| 22 |
+
## Project
|
|
|
|
|
|
|
| 23 |
|
| 24 |
+
- Paper: https://arxiv.org/abs/2608.11521
|
| 25 |
+
- Code: https://github.com/ChushanZhang/RIFT
|
| 26 |
+
- Dataset: https://huggingface.co/datasets/yuanty/LIBERO-fastwam
|
| 27 |
|
| 28 |
+
## Files
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
|
| 30 |
+
```text
|
| 31 |
+
rift_step021700.pt
|
| 32 |
+
dataset_stats.json
|
| 33 |
+
config.yaml
|
| 34 |
+
```
|
| 35 |
|
| 36 |
+
- `rift_step021700.pt`: released RIFT checkpoint for LIBERO
|
| 37 |
+
- `dataset_stats.json`: normalization statistics used by the checkpoint
|
| 38 |
+
- `config.yaml`: model configuration
|
| 39 |
|
| 40 |
+
Keep `dataset_stats.json` with the checkpoint during evaluation.
|
|
|
|
|
|
|
|
|
|
|
|
|
| 41 |
|
| 42 |
+
`rift_step021700.pt` uses PyTorch serialization. Download it only from this
|
| 43 |
+
repository and load it with `weights_only=True`.
|
|
|
|
|
|
|
|
|
|
| 44 |
|
| 45 |
## Download
|
| 46 |
|
| 47 |
```bash
|
| 48 |
+
pip install -U huggingface_hub
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 49 |
|
| 50 |
+
hf download PoopBear/RIFT \
|
| 51 |
+
rift_step021700.pt \
|
| 52 |
+
dataset_stats.json \
|
| 53 |
+
config.yaml \
|
| 54 |
+
--local-dir ./checkpoints/rift
|
| 55 |
```
|
| 56 |
|
| 57 |
+
Evaluation commands are provided in the
|
| 58 |
+
[RIFT repository](https://github.com/ChushanZhang/RIFT#benchmark-evaluation).
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 59 |
|
| 60 |
+
## Citation
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 61 |
|
| 62 |
+
```bibtex
|
| 63 |
+
@article{zhang2026rift,
|
| 64 |
+
title={Keep the Future, Drop the Rollout: RIFT for World Action Models},
|
| 65 |
+
author={Zhang, Chushan and Tong, Jinguang and Li, Xuesong and Wang, Yikai and Li, Hongdong},
|
| 66 |
+
journal={arXiv preprint arXiv:2608.11521},
|
| 67 |
+
year={2026}
|
| 68 |
+
}
|
| 69 |
```
|
| 70 |
|
| 71 |
+
## License
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 72 |
|
| 73 |
+
The checkpoint is distributed under the Apache License 2.0. RIFT source code
|
| 74 |
+
is distributed separately under the MIT License. See `NOTICE` for attribution.
|