Skip to content

Article

AWLS: Advantage-Weighted Latent Steering for One-Step Generative Robotic Manipulation Policies (CoRL 2026)

arXiv cs.RO

A CoRL 2026 paper proposes Advantage-Weighted Latent Steering (AWLS), an offline post-training method that improves frozen one-step generative manipulation policies without updating their decoders.

Part of the event: AWLS: An Offline Post-Training Method for Frozen One-Step Robot Manipulation Policies

This page provides a summary and source link only.