Views
No views yet
robomimic/ — Can, Square, Transport, Tool Hang (available now)libero/ — coming soon| Task | Demos | Notes |
|---|---|---|
can | 15 | policy + dynamics model |
square | 30 | policy + dynamics model |
transport | 10 | policy + dynamics model |
tool_hang | 80 | policy + dynamics model |
robomimic/diffusion_policy/<task>/<variant>/
├── checkpoints/<epoch>.ckpt # policy weights (~4.3–5.3 GB)
├── normalizer.pth # observation/action normalizer
└── .hydra/ # full training config
robomimic/dyn_model/<task>/
├── checkpoints/model_<epoch>.pth # visual dynamics model (~0.5–1.5 GB)
├── normalizer.pth
└── hydra.yaml # training config880.ckpt) is the training epoch.| Variant | What it is |
|---|---|
base_policy | Diffusion policy trained on the original demonstrations only. |
ReGuide-FS | Retrained from scratch on demos + guided rollouts. |
ReGuide-FT-iteration1 | Base checkpoint fine-tuned on demos + guided rollouts (first iteration). |
ReGuide-FT-iteration2 | Second fine-tuning iteration on freshly collected guided rollouts. |
ReGuide-FS-FT | ReGuide-FT applied on top of a ReGuide-FS policy (best on Can/Square/Transport). |
.hydra/ for policies,
hydra.yaml for dynamics models), which records the exact hyperparameters used. Paths
inside the configs are relative.1@article{lin2026reguide,
2 title = {ReGuide: From Test-Time Guidance to Self-Improving Diffusion Policies},
3 author = {Lin, Tzu-Hsiang and Shakkottai, Srinivas and Kalathil, Dileep and Kumar, P. R.},
4 journal = {arXiv preprint arXiv:2606.28939},
5 year = {2026},
6 eprint = {2606.28939},
7 archivePrefix = {arXiv},
8 primaryClass = {cs.LG},
9 url = {https://arxiv.org/abs/2606.28939}
10}