DreamGen trains robot policies on synthetic trajectories from adapted video world models, enabling a humanoid robot to perform 22 new behaviors in seen and unseen environments from a single pick-and-place teleoperation dataset.
Physics-driven data generation for contact-rich manipulation via trajectory optimization
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
verdicts
UNVERDICTED 8roles
background 1polarities
background 1representative citing papers
DO AS I DO reconstructs and retargets hand-object interactions from in-the-wild monocular RGB videos to produce dexterous robot manipulation trajectories, outperforming prior methods on ground-truth and online video datasets.
A first-order Sobolev loss for diffusion policies enables warm-starting trajectory optimization solvers with 2×–20× speedup and fewer diffusion steps, using very few training trajectories.
GR00T N1 is a new open VLA foundation model for humanoid robots that outperforms imitation learning baselines in simulation and shows strong performance on real-world bimanual manipulation tasks.
PhysisForcing applies trajectory and relational alignment losses to DiT features in video models, improving physical plausibility on R-Bench, PAI-Bench, and EZS-Bench while raising closed-loop robotic success rates from 16% to 24%.
HumanoidMimicGen automatically generates large loco-manipulation datasets from few source demonstrations using whole-body planning, enabling visuomotor policies that outperform real-data-only training by 20% on a new nine-task benchmark.
Framework generates force-informed sim data from one demo to train compliant visuomotor flow matching policies, showing reliable contact on real-robot block flipping and bi-manual tasks.
MinInter improves imitation learning data augmentation by choosing source demonstrations that minimize inserted interpolation segments, yielding higher success rates on 12 MimicGen tasks.
citing papers explorer
-
DreamGen: Unlocking Generalization in Robot Learning through Video World Models
DreamGen trains robot policies on synthetic trajectories from adapted video world models, enabling a humanoid robot to perform 22 new behaviors in seen and unseen environments from a single pick-and-place teleoperation dataset.
-
Do as I Do: Dexterous Manipulation Data from Everyday Human Videos
DO AS I DO reconstructs and retargets hand-object interactions from in-the-wild monocular RGB videos to produce dexterous robot manipulation trajectories, outperforming prior methods on ground-truth and online video datasets.
-
Accelerating trajectory optimization with Sobolev-trained diffusion policies
A first-order Sobolev loss for diffusion policies enables warm-starting trajectory optimization solvers with 2×–20× speedup and fewer diffusion steps, using very few training trajectories.
-
GR00T N1: An Open Foundation Model for Generalist Humanoid Robots
GR00T N1 is a new open VLA foundation model for humanoid robots that outperforms imitation learning baselines in simulation and shows strong performance on real-world bimanual manipulation tasks.
-
PhysisForcing: Physics Reinforced World Simulator for Robotic Manipulation
PhysisForcing applies trajectory and relational alignment losses to DiT features in video models, improving physical plausibility on R-Bench, PAI-Bench, and EZS-Bench while raising closed-loop robotic success rates from 16% to 24%.
-
HumanoidMimicGen: Data Generation for Loco-Manipulation via Whole-Body Planning
HumanoidMimicGen automatically generates large loco-manipulation datasets from few source demonstrations using whole-body planning, enabling visuomotor policies that outperform real-data-only training by 20% on a new nine-task benchmark.
-
Flow with the Force Field: Learning 3D Compliant Flow Matching Policies from Force and Demonstration-Guided Simulation Data
Framework generates force-informed sim data from one demo to train compliant visuomotor flow matching policies, showing reliable contact on real-robot block flipping and bi-manual tasks.
-
MinInter: Minimizing Trajectory Interpolation During Data Augmentation for Imitation Learning
MinInter improves imitation learning data augmentation by choosing source demonstrations that minimize inserted interpolation segments, yielding higher success rates on 12 MimicGen tasks.