haohw
/

behavior-uncloning-baseline

Model card Files Files and versions

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

Behavior Uncloning - Baseline DiffusionPolicy

Baseline DiffusionPolicy trained on all LIBERO-10 episodes (379 episodes, ~101K frames).

Checkpoints

step_025000/ - 25K steps, 76% overall success rate
step_050000/ - 50K steps, 76% overall success rate (final)

Training Config

Model: DiffusionPolicy (267M params)
Dataset: HuggingFaceVLA/libero (LIBERO-10 subset)
Effective batch size: 64 x 8 GPUs = 512
Training time: ~20 hours on 8xA100-80GB
Framework: LeRobot v0.4.4

Per-Task Success Rates (step 50K)

T0	T1	T2	T3	T4	T5	T6	T7	T8	T9
10%	60%	80%	100%	80%	100%	60%	100%	70%	100%

Usage

from lerobot.policies.diffusion.modeling_diffusion import DiffusionPolicy
policy = DiffusionPolicy.from_pretrained("haohw/behavior-uncloning-baseline", subfolder="step_050000")

Project

Part of the behavior-uncloning project — machine unlearning for robot manipulation policies.

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support