Handheld demonstrations improve robot policies without repeated robot execution

HIL-UMI targets out-of-distribution states and useful task segments while humans demonstrate with a Universal Manipulation Interface.

Top University
Zimu Han · Yiming Zeng · Jiyao Zhang · Zihao Zhao · Yuanfei Wang · Yixiang Jin · +6 more

Xi’an Jiaotong University · National Key Laboratory for Multimedia Information Processing · Peking University · PrimeBot · Center on Frontier Computing Studies

Research Digest··2 min read
Han et al.

The authors built HIL-UMI around handheld Universal Manipulation Interface demonstrations.

Why this paper

From Peking University and 5 others · Part of Credit Assignment in Agentic RL, now 17 papers

In one line

HIL-UMI improves VLA robot policies via human-in-the-loop data collection without physical robot deployment.

What we could check

  • ·No code link found
  • ·No weights link found
  • ·No dataset link found
  • ·No compute details found
  • ·No stated limitations found
  • ·No benchmark numbers found

Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.

§
newspaper

Research Digest

Articles published under the Zotpaper byline are synthesized from multiple source publications by our AI editor and reviewed by our editorial process. Each story combines reporting from credible outlets to give readers a balanced, comprehensive view.