---
title: "REGRIND: A Minimalist Retargeting-Guided Reinforcement Learning Recipe for Dexterous Manipulation"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2607.11874v1"
permalink: "https://feed.astralor.com/items/itm_0a441b05c2deda5a4e3f68a2"
published: "2026-07-13T17:56:08.000Z"
score: 78
channel: "ai"
tags: ["REGRIND","Simulation-to-real transfer","Reinforcement Learning","Dexterous Manipulation"]
---

# REGRIND: A Minimalist Retargeting-Guided Reinforcement Learning Recipe for Dexterous Manipulation

## AI commentary

> This paper brings the successful 'retargeting+RL' recipe from humanoid whole-body control to dexterous manipulation, training transferable policies from a single human demo—great for both labs and real-world deployment. If you're struggling with sim-to-real gaps, the system identification and domain randomization analysis here offer practical takeaways.

## At a glance

A new arXiv paper presents REGRIND, a minimalist pipeline that learns dexterous manipulation policies from a single human demonstration. It retargets human hand-object motion to robot kinematic references preserving spatial and contact relationships, trains a residual RL policy in simulation to track object-centric keypoints, and transfers zero-shot to hardware with careful system identification. The resulting policies produce fluid, human-like behavior on multi-fingered hands for contact-rich tool-use tasks like scissors and screwdriver. Systematic hardware experiments identify key factors governing sim-to-real transfer.

## Key points

- Proposes REGRIND, a minimalist pipeline learning dexterous manipulation from a single human demo.
- Uses retargeting to preserve hand-object spatial/contact relations, with residual RL and system ID for zero-shot transfer.
- Validated on multi-fingered hands for contact-rich tasks (scissors, screwdriver), achieving human-like fluidity.

## Full text

Recent work in humanoid whole-body control has found success with a simple recipe: retarget human motion to robot kinematic references, then train policies via reinforcement learning (RL) to track them. But how does this recipe transfer to dexterous manipulation? The answer is not obvious, as manipulation involves complex, contact-rich dynamics and requires delicate regulation of contact modes and forces. We present REGRIND, a minimalist retargeting-guided RL pipeline that learns dexterous manipulation policies from a single human demonstration. REGRIND retargets human hand-object motion to a robot reference that preserves hand-object spatial and contact relationships, trains a residual RL policy in simulation to track object-centric keypoints along that reference, and transfers the resulting policy zero-shot to hardware with careful system identification. The resulting policies produce fluid, human-like behavior on two different multi-fingered hands across contact-rich tool-use tasks, including operating a pair of scissors and turning a screwdriver. Through systematic hardware experiments, we identify and analyze the key factors that govern sim-to-real transfer in dexterous manipulation, offering practical guidance for retargeting-based learning in contact-rich settings. Videos and code are available at https://yunhaifeng.com/REGRIND.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_0a441b05c2deda5a4e3f68a2