Sunghwan Kim

PhD student @UCSanDiego

[Sunghwan Kim @sssssshwan]

Introducing SERF: a spatiotemporal environment and robot feature map for long-horizon mobile manipulation.

We demonstrate that conditioning a mobile manipulation policy on a SERF map enables long-horizon reasoning.

[Sunghwan Kim @sssssshwan]

@astridwilde1 Thanks for sharing our work Astrid!

[Sunghwan Kim @sssssshwan]

@xyz2maureen Thanks for sharing our work Xueyan!

[Sunghwan Kim @sssssshwan]

@chris_j_paxton Thanks for sharing our work Chris!

[Sunghwan Kim @sssssshwan]

It was a pleasure working with this amazing team: @Byeonghyun_Pak (co-first), Kehan Long, Yulun Tian, and @natanaso!

[Sunghwan Kim @sssssshwan]

Introducing SERF: a spatiotemporal environment and robot feature map for long-horizon mobile manipulation.

We demonstrate that conditioning a mobile manipulation policy on a SERF map enables long-horizon reasoning.

[Sunghwan Kim @sssssshwan]

Check out the project page, paper, and code for more details and results.

Project: https://t.co/s4zUvTWp58
Paper: https://t.co/fTUgt14lz2
Mapping code: https://t.co/QLHvRrhazW
VLA code: https://t.co/JXBVptUMQN

9/n

[Junwon Seo @Junwon__Seo]

Video world model imaginations🌎💭can miss critical but plausible outcomes of robot actions.

Introducing 𝙎𝙩𝙧𝙚𝙨𝙨𝘿𝙧𝙚𝙖𝙢: inference-time steering for video WMs, imagining plausible✅, high-impact⚠️ futures for 𝙧𝙤𝙗𝙪𝙨𝙩🛡️ policy evaluation and improvement.

(1/15)