RLBench

RLBench is a benchmark that contains a diverse set of robotic manipulation tasks and is used to evaluate the performance of Vision-Language Models in detecting and reasoning about failures in robotic manipulation.

Papers using RLBench (7)

PointACT: Vision-Language-Action Models with Multi-Scale Point-Action Interaction2026VLA-Pro: Cross-Task Procedural Memory Transfer for Vision-Language-Action Models2026Scaling Cross-Environment Failure Reasoning Data for Vision-Language Robotic Manipulation2025Audio-VLA: Adding Contact Audio Perception to Vision-Language-Action Model for Robotic Manipulation2025Learning To See And Act: Task-aware Virtual View Exploration For Robotic Manipulation2025Mini Diffuser: Fast Multi-task Diffusion Policy Training Using Two-level Mini-batches2025Large Pre-trained Models For Bimanual Manipulation In 3D2025