# ReDex boosts real‑world success of simulation‑trained hand policies

> The new framework lifts object‑flipping success from 14% to 86% and screwdriver rotation progress from 26% to 95% on real robots.

Oossa · 2026-10-07 · https://oossa.com/en/redex-boosts-real-world-success-of-simulation-trained-hand-policies

Researchers led by Jinzhou Li released ReDex, a method for adapting dexterous hand policies trained in simulation to real hardware. The system lets a human gently correct finger contacts while the original policy runs the other fingers. Rollouts that mix the base policy, human‑guided finger moves, and fingertip force data are used to train a new force‑conditioned policy via behavior cloning. In tests on two contact‑rich tasks, ReDex raised object‑flipping success from 14% to 86% and screwdriver rotation progress from 26% to 95.3%.

## The facts

- Object flipping success rose from 14% to 86% across two objects.
- Screwdriver rotation progress improved from 26.0% to 95.3% across three objects.

## Why it matters

It shows a practical way to turn simulation‑trained hand skills into reliable real‑world robot actions with minimal human correction.

## Sources & references

1. [ReDex: Repairing Sim-to-Real Dexterous Policies by Finger-Level Compliant Interaction](https://arxiv.org/abs/2610.07525) – arXiv cs.RO, 2026-10-07

Last updated: 2026-10-07