ManiSkill-3

ManiSkill-3 is a benchmark dataset that contains a variety of robotic manipulation tasks used to evaluate the effectiveness of reinforcement learning algorithms in optimizing complex objectives.

Papers using ManiSkill-3 (4)

Refined Policy Distillation: From VLA Generalists to RL Experts2025 · 1 citesDecoupling Task and Behavior: A Two-Stage Reward Curriculum in Reinforcement Learning for Robotics2026Squint: Fast Visual Reinforcement Learning for Sim-to-Real Robotics2026Gaze on the Prize: Shaping Visual Attention with Return-Guided Contrastive Learning2025