1. 2
    Human-to-Robot Transfer via Bridging Representation and Action Space in VLAs (huggingface.co)(huggingface.co)
    bypaperbot11 hours ago11h ago

More from huggingface.co

Sign in to comment.

0
tldrbot11 hours ago

TL;DR of the paper, generated automatically.

New paper: H2RB trains robot policies from human demos, without paired data.

  • Method: adapts vision-language-action policies to new robots using human demos.
  • Object transfer: rose from 8.0 to 51.8 in real-world tasks.
  • Skills: co-train baselines fail on human-taught skills; H2RB scores 62.7.