RLBench

RLBench is a dataset and benchmark that contains a variety of continuous manipulation tasks used to evaluate robotic control and planning in cloud-robotic systems.

Papers using RLBench (40)

Demo-JEPA: Joint-Embedding Predictive Architecture for One-shot Cross-Embodiment Imitation2026Learning Structural Latent Points for Efficient Visual Representations in Robotic Manipulation2026PointACT: Vision-Language-Action Models with Multi-Scale Point-Action Interaction2026VLA-Pro: Cross-Task Procedural Memory Transfer for Vision-Language-Action Models2026Mask World Model: Predicting What Matters for Robust Robot Policy Learning2026Action Images: End-to-End Policy Learning via Multiview Video Generation2026TGM-VLA: Task-Guided Mixup for Sampling-Efficient and Robust Robotic Manipulation2026Hyperbolic Multiview Pretraining for Robotic Manipulation2026ST-VLA: Enabling 4D-Aware Spatiotemporal Understanding for General Robot Manipulation2026Speculative Policy Orchestration: A Latency-Resilient Framework for Cloud-Robotic Manipulation2026Cortical Policy: A Dual-Stream View Transformer for Robotic Manipulation2026GSR: Learning Structured Reasoning for Embodied Manipulation2026Learning Geometrically-Grounded 3D Visual Representations for View-Generalizable Robotic Manipulation2026Scaling Cross-Environment Failure Reasoning Data for Vision-Language Robotic Manipulation2025VERM: Leveraging Foundation Models to Create a Virtual Eye for Efficient 3D Robotic Manipulation2025SpatialActor: Exploring Disentangled Spatial Representations for Robust Robotic Manipulation2025Towards Reliable Code-as-Policies: A Neuro-Symbolic Framework for Embodied Task Planning2025DynaRend: Learning 3D Dynamics via Masked Future Rendering for Robotic Manipulation2025Large Pre-Trained Models for Bimanual Manipulation in 3D2025Memory Transfer Planning: LLM-driven Context-Aware Code Adaptation for Robot Manipulation2025Multi-Modal Manipulation via Multi-Modal Policy Consensus2025ViReSkill: Vision-Grounded Replanning with Skill Memory for LLM-Based Planning in Lifelong Robot Learning2025Learning to See and Act: Task-Aware Virtual View Exploration for Robotic Manipulation2025Actor-Critic for Continuous Action Chunks: A Reinforcement Learning Framework for Long-Horizon Robotic Manipulation with Sparse Reward2025ADPro: a Test-time Adaptive Diffusion Policy via Manifold-constrained Denoising and Task-aware Initialization for Robotic Manipulation2025Ego-centric Predictive Model Conditioned on Hand Trajectories2025FMimic: Foundation Models are Fine-grained Action Learners from Human Videos2025Learning Video Generation for Robotic Manipulation with Collaborative Trajectory Control2025BridgeVLA: Input-Output Alignment for Efficient 3D Manipulation Learning with Vision-Language Models2025Chain-of-Action: Trajectory Autoregressive Modeling for Robotic Manipulation2025FlowRAM: Grounding Flow Matching Policy with Region-Aware Mamba Framework for Robotic Manipulation2025MinD: Learning A Dual-System World Model for Real-Time Planning and Implicit Risk Analysis2025RoboPearls: Editable Video Simulation for Robot Manipulation2025The Unreasonable Effectiveness of Discrete-Time Gaussian Process Mixtures for Robot Policy Learning2025GPA-RAM: Grasp-Pretraining Augmented Robotic Attention Mamba for Spatial Task Learning2025NeSyC: A Neuro-symbolic Continual Learner For Complex Embodied Tasks In Open Domains2025GraspCorrect: Robotic Grasp Correction via Vision-Language Model-Guided Feedback2025Robotic Programmer: Video Instructed Policy Code Generation for Robotic Manipulation2025RoboHorizon: An LLM-Assisted Multi-View World Model for Long-Horizon Robotic Manipulation2025SAM2Act: Integrating Visual Foundation Model with A Memory Architecture for Robotic Manipulation2025