Article

From Grasping to Dexterous Manipulation: Large-Scale Grasp Pretraining for Dexterous Manipulation

The research team proposes pretraining the low-level controller of a hierarchical imitation learning framework with large-scale dexterous grasp data, then fine-tuning it on downstream task demonstrations to support functional dexterity in articulated tool use. They built a 355k-trajectory grasp-pretraining dataset from large-scale dexterous grasp annotations and introduced DexCraft, a simulation benchmark with six articulated tool-use tasks. In simulation and real-world experiments, the method outperforms end-to-end diffusion policy baselines and hierarchical policies trained from scratch, improving real-world full-task success by 33.3 percentage points over DP3.

Part of the event: Grasp Pretraining Enhances Hierarchical Imitation Learning for Dexterous Manipulation

This page provides a summary and source link only.