LIBERO is a new benchmark for lifelong robot learning that evaluates transfer of declarative, procedural, and mixed knowledge across 130 manipulation tasks with provided demonstration data.
hub
What Matters in Learning from Offline Human Demonstrations for Robot Manipulation
27 Pith papers cite this work. Polarity classification is still indexing.
27
Pith papers citing it
abstract
Imitating human demonstrations is a promising approach to endow robots with various manipulation capabilities. While recent advances have been made in imitation learning and batch (offline) reinforcement learning, a lack of open-source human datasets and reproducible learning methods make assessing the state of the field difficult. In this paper, we conduct an extensive study of six offline learning algorithms for robot manipulation on five simulated and three real-world multi-stage manipulation tasks of varying complexity, and with datasets of varying quality. Our study analyzes the most critical challenges when learning from offline human data for manipulation. Based on the study, we derive a series of lessons including the sensitivity to different algorithmic design choices, the dependence on the quality of the demonstrations, and the variability based on the stopping criteria due to the different objectives in training and evaluation. We also highlight opportunities for learning from human datasets, such as the ability to learn proficient policies on challenging, multi-stage tasks beyond the scope of current reinforcement learning methods, and the ability to easily scale to natural, real-world manipulation scenarios where only raw sensory signals are available. We have open-sourced our datasets and all algorithm implementations to facilitate future research and fair comparisons in learning from human demonstration data. Codebase, datasets, trained models, and more available at https://arise-initiative.github.io/robomimic-web/
hub tools
citation-role summary
background 1
dataset 1
citation-polarity summary
polarities
background 2representative citing papers
Flow map policies enable fast one-step inference for flow-based RL policies, and FMQ provides an optimal closed-form Q-guided target for offline-to-online adaptation under trust-region constraints, achieving SOTA performance.
A liveness-based Bellman operator enables conservative offline policy evaluation for manipulation tasks by encoding task progression and reducing truncation bias from finite horizons.
OmniNavBench is a unified benchmark for general-purpose navigation featuring composite multi-skill instructions, support for humanoid, quadrupedal and wheeled robots, and 1779 human teleoperated trajectories across 170 environments.
ScoRe-Flow achieves decoupled mean-variance control in stochastic flow matching by deriving a closed-form score for drift modulation plus learned variance, yielding faster RL convergence and higher success rates on locomotion and manipulation benchmarks.
ReV is a referring-aware visuomotor policy using coupled diffusion heads for real-time trajectory replanning in robotic manipulation, trained solely via targeted perturbations to expert demonstrations and achieving higher success rates in simulated and real tasks.
A collaborative dataset spanning 22 robots and 527 skills enables RT-X models that transfer capabilities across different robot embodiments.
SID achieves approximately 90% success on six real-world manipulation tasks with only two demonstrations under out-of-distribution initializations, with less than 10% performance drop under distractors and disturbances.
StereoPolicy fuses stereo image pairs via a Stereo Transformer on pretrained 2D encoders to boost robotic manipulation policies, showing gains over monocular, RGB-D, point cloud, and multi-view methods in simulations and real-robot tests.
DexSynRefine synthesizes HOI motions with an extended manifold method, refines them via task-space residual RL, and adapts for sim-to-real transfer, outperforming kinematic retargeting by 50-70 percentage points on five dexterous tasks.
LQL turns n-step action-sequence lower bounds into a practical hinge-loss stabilizer for off-policy Q-learning without extra networks or forward passes.
Q2RL extracts Q-functions from BC policies via minimal interactions and applies Q-gating to enable stable offline-to-online RL, outperforming baselines on manipulation benchmarks and achieving up to 100% success on-robot.
A multi-agent RL high-level planner outputs task-space velocities that a GPU-parallel QP low-level controller converts to joint velocities while enforcing limits and collisions, yielding robust sim-to-real dexterous grasping with zero-shot steerability.
OGPO is a sample-efficient off-policy method for full finetuning of generative control policies that reaches SOTA on robotic manipulation tasks and can recover from poor behavior-cloning initializations without expert data.
Hydra-DP3 achieves SOTA visuomotor performance with under 1% of prior 3D diffusion policy parameters by using frequency analysis to justify a lightweight decoder and two-step DDIM inference.
Power spectral density of trajectories ranks demonstration quality for imitation learning, enabling rollout-free curation that improves fine-tuned policy success.
RINSE scores robot demonstration trajectories for smoothness via SAL and TED metrics to curate higher-quality data for behavioral cloning, improving success rates with less data on benchmarks and real robots.
VADF adds an Adaptive Loss Network for hard-negative training sampling and a Hierarchical Vision Task Segmenter for adaptive noise scheduling during inference to speed convergence and reduce timeouts in diffusion robotic policies.
WARPED synthesizes realistic wrist-view observations from monocular egocentric human videos via foundation models, hand-object tracking, retargeting, and Gaussian Splatting to train visuomotor policies that match teleoperation success rates on five tabletop tasks with 5-8x less collection effort.
Unified World Models couple video and action diffusion inside one transformer with independent timesteps, enabling pretraining on heterogeneous robot datasets that include action-free video and producing more generalizable policies than imitation learning alone.
UVA learns a joint video-action latent representation with decoupled diffusion decoding heads, enabling a single model to perform accurate fast policy learning, forward/inverse dynamics, and video generation without performance loss versus task-specific methods.
DROID is a new 76k-trajectory in-the-wild robot manipulation dataset spanning 564 scenes and 84 tasks that improves policy performance and generalization when used for training.
Execution guarantee certifies safe regions for IL policies via view synthesis and set invariance so that maximum task success is assured from within those regions even under small execution changes.
GMP selectively activates and represents memory via a gate and lightweight cross-attention, yielding 30.1% higher success on non-Markovian robotic tasks while staying competitive on Markovian ones.
citing papers explorer
-
LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning
LIBERO is a new benchmark for lifelong robot learning that evaluates transfer of declarative, procedural, and mixed knowledge across 130 manipulation tasks with provided demonstration data.
-
Aligning Flow Map Policies with Optimal Q-Guidance
Flow map policies enable fast one-step inference for flow-based RL policies, and FMQ provides an optimal closed-form Q-guided target for offline-to-online adaptation under trust-region constraints, achieving SOTA performance.
-
Offline Policy Evaluation for Manipulation Policies via Discounted Liveness Formulation
A liveness-based Bellman operator enables conservative offline policy evaluation for manipulation tasks by encoding task progression and reducing truncation bias from finite horizons.
-
Beyond Isolation: A Unified Benchmark for General-Purpose Navigation
OmniNavBench is a unified benchmark for general-purpose navigation featuring composite multi-skill instructions, support for humanoid, quadrupedal and wheeled robots, and 1779 human teleoperated trajectories across 170 environments.
-
ScoRe-Flow: Complete Distributional Control via Score-Based Reinforcement Learning for Flow Matching
ScoRe-Flow achieves decoupled mean-variance control in stochastic flow matching by deriving a closed-form score for drift modulation plus learned variance, yielding faster RL convergence and higher success rates on locomotion and manipulation benchmarks.
-
Referring-Aware Visuomotor Policy Learning for Closed-Loop Manipulation
ReV is a referring-aware visuomotor policy using coupled diffusion heads for real-time trajectory replanning in robotic manipulation, trained solely via targeted perturbations to expert demonstrations and achieving higher success rates in simulated and real tasks.
-
Open X-Embodiment: Robotic Learning Datasets and RT-X Models
A collaborative dataset spanning 22 robots and 527 skills enables RT-X models that transfer capabilities across different robot embodiments.
-
SID: Sliding into Distribution for Robust Few-Demonstration Manipulation
SID achieves approximately 90% success on six real-world manipulation tasks with only two demonstrations under out-of-distribution initializations, with less than 10% performance drop under distractors and disturbances.
-
StereoPolicy: Improving Robotic Manipulation Policies via Stereo Perception
StereoPolicy fuses stereo image pairs via a Stereo Transformer on pretrained 2D encoders to boost robotic manipulation policies, showing gains over monocular, RGB-D, point cloud, and multi-view methods in simulations and real-robot tests.
-
DexSynRefine: Synthesizing and Refining Human-Object Interaction Motion for Physically Feasible Dexterous Robot Actions
DexSynRefine synthesizes HOI motions with an extended manifold method, refines them via task-space residual RL, and adapts for sim-to-real transfer, outperforming kinematic retargeting by 50-70 percentage points on five dexterous tasks.
-
Long-Horizon Q-Learning: Accurate Value Learning via n-Step Inequalities
LQL turns n-step action-sequence lower bounds into a practical hinge-loss stabilizer for off-policy Q-learning without extra networks or forward passes.
-
When Life Gives You BC, Make Q-functions: Extracting Q-values from Behavior Cloning for On-Robot Reinforcement Learning
Q2RL extracts Q-functions from BC policies via minimal interactions and applies Q-gating to enable stable offline-to-online RL, outperforming baselines on manipulation benchmarks and achieving up to 100% success on-robot.
-
Learning Reactive Dexterous Grasping via Hierarchical Task-Space RL Planning and Joint-Space QP Control
A multi-agent RL high-level planner outputs task-space velocities that a GPU-parallel QP low-level controller converts to joint velocities while enforcing limits and collisions, yielding robust sim-to-real dexterous grasping with zero-shot steerability.
-
OGPO: Sample Efficient Full-Finetuning of Generative Control Policies
OGPO is a sample-efficient off-policy method for full finetuning of generative control policies that reaches SOTA on robotic manipulation tasks and can recover from poor behavior-cloning initializations without expert data.
-
Hydra-DP3: Frequency-Aware Right-Sizing of 3D Diffusion Policies for Visuomotor Control
Hydra-DP3 achieves SOTA visuomotor performance with under 1% of prior 3D diffusion policy parameters by using frequency analysis to justify a lightweight decoder and two-step DDIM inference.
-
An Efficient Metric for Data Quality Measurement in Imitation Learning
Power spectral density of trajectories ranks demonstration quality for imitation learning, enabling rollout-free curation that improves fine-tuned policy success.
-
Learning from the Best: Smoothness-Driven Metrics for Data Quality in Imitation Learning
RINSE scores robot demonstration trajectories for smoothness via SAL and TED metrics to curate higher-quality data for behavioral cloning, improving success rates with less data on benchmarks and real robots.
-
VADF: Vision-Adaptive Diffusion Policy Framework for Efficient Robotic Manipulation
VADF adds an Adaptive Loss Network for hard-negative training sampling and a Hierarchical Vision Task Segmenter for adaptive noise scheduling during inference to speed convergence and reduce timeouts in diffusion robotic policies.
-
WARPED: Wrist-Aligned Rendering for Robot Policy Learning from Egocentric Human Demonstrations
WARPED synthesizes realistic wrist-view observations from monocular egocentric human videos via foundation models, hand-object tracking, retargeting, and Gaussian Splatting to train visuomotor policies that match teleoperation success rates on five tabletop tasks with 5-8x less collection effort.
-
Unified World Models: Coupling Video and Action Diffusion for Pretraining on Large Robotic Datasets
Unified World Models couple video and action diffusion inside one transformer with independent timesteps, enabling pretraining on heterogeneous robot datasets that include action-free video and producing more generalizable policies than imitation learning alone.
-
Unified Video Action Model
UVA learns a joint video-action latent representation with decoupled diffusion decoding heads, enabling a single model to perform accurate fast policy learning, forward/inverse dynamics, and video generation without performance loss versus task-specific methods.
-
DROID: A Large-Scale In-The-Wild Robot Manipulation Dataset
DROID is a new 76k-trajectory in-the-wild robot manipulation dataset spanning 564 scenes and 84 tasks that improves policy performance and generalization when used for training.
-
To Do or Not to Do: Ensuring the Safety of Visuomotor Policies Learned from Demonstrations
Execution guarantee certifies safe regions for IL policies via view synthesis and set invariance so that maximum task success is assured from within those regions even under small execution changes.
-
Gated Memory Policy
GMP selectively activates and represents memory via a gate and lightweight cross-attention, yielding 30.1% higher success on non-Markovian robotic tasks while staying competitive on Markovian ones.
-
Efficient Hierarchical Implicit Flow Q-learning for Offline Goal-conditioned Reinforcement Learning
Proposes mean flow policies and LeJEPA loss to overcome Gaussian policy limits and weak subgoal generation in hierarchical offline GCRL, reporting strong results on OGBench state and pixel tasks.
-
From Video to Control: A Survey of Learning Manipulation Interfaces from Temporal Visual Data
A survey introduces an interface-centric taxonomy for video-to-control methods in robotic manipulation and identifies the robotics integration layer as the central open challenge.
-
World Action Models: The Next Frontier in Embodied AI
The paper introduces World Action Models as a new paradigm unifying predictive world modeling with action generation in embodied foundation models and provides a taxonomy of existing approaches.