HQL balances one-step and multi-step returns for long horizons
A multi-step target is kept only when a dataset action beats best-of-N behavior-cloned actions.
Stories per day · 30 days
Relatedrobotics2demonstration-curation1force-control1humanoid1imitation-learning1loco-manipulation1offline-rl1ogbench1q-learning1reinforcement-learning1robomimic1
5 stories, newest first · one card per story
Stories tagged #diffusion-policy
CuraEval finds demo curation rarely beats full robot training data
On robomimic, no curated subset significantly beat full-data training after Holm correction, though intervals still allow gains.
FLORE learns force-reactive humanoid loco-manipulation from handheld demos
On three contact-rich robot tasks the force-reactive policy succeeds in 80% of rollouts versus 26.7% for position-only.
Learning Foresight without Explicit Trajectories for 3D Diffusion Policies
Movement Trend Guidance adds latent foresight to 3D diffusion policies, improving DP3 across RoboTwin2.0, LIBERO-40, and real-robot tasks with only 3.52% more parameters.
Attention-DP3: Spatially Object-aware 3D Diffusion Policy via Geometry-aligned Attentional Conditioning
Attention-DP3 adds spatially object-aware attentional conditioning to 3D diffusion policies, improving robotic manipulation by up to 31% under heavy clutter.