ManiSkill-3

'ManiSkill-3' is a benchmark dataset that evaluates robotic manipulation tasks using egocentric visual perception, focusing on the challenges of occlusion, motion blur, and partial observability in dynamic environments.

Papers using ManiSkill-3 (7)

Feat2Go: Visual Feature-Grounded Value Estimation for Embodied Reinforcement Learning2026Fourier Features Let Agents Learn High Precision Policies with Imitation Learning2026Multi-Slot Attention with State Guidance for Egocentric Robotic Manipulation2026Decoupling Task And Behavior: A Two-stage Reward Curriculum In Reinforcement Learning For Robotics2026Demonstration-Free Robotic Control via LLM Agents2026STARE-VLA: Progressive Stage-Aware Reinforcement for Fine-Tuning Vision-Language-Action Models2025Merging and Disentangling Views in Visual Reinforcement Learning for Robotic Manipulation2025