A Smooth Sea Never Made a Skilled SAILOR: Robust Imitation via Learning to Search
Paper
•
2506.05294
•
Published
⛵ SAILOR: Searching Across Imagined Latents Online for Recovery
Model checkpoints for the robomimic tasks for all different expert dataset sizes mentioned in the paper. Files located under structure
$TASK_NAME/ # e.g. robomimic__lift, robomimic__can, robomimic__square
└── demos_$DEMOS/ # e.g. demos_5, demos_10, demos_50
└── seed_$SEED/ # e.g. seed_0
├── latest_base_policy.pt
└── latest_residual_checkpoint.pt
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support