UMI-WM: Interactive World Modeling for Bimanual Manipulation from Wrist Cameras
Overview
UMI-WM is an action-conditioned bimanual world model that jointly generates both wrist views from observation history, per-arm relative camera poses, and gripper widths, with cross-view consistency, action following, and visual fidelity. The model outperforms baselines in action following and generates paired wrist videos 6.9 times faster than the Teacher model.
From arXiv Robotics — research abstracts
Event evidence and corrections
0 attributed source owners. Ownership does not establish independent confirmation. Quantities are reported separately and are never added together.
Reported quantity · generated videos: 6.9 other · Basis not reported
“following, while the four-step Student generates paired wrist videos 6.9 times faster than the Teacher.”
Exact source · revision 1Source owner not reported
Reported quantity · denoising steps: 4 other · Basis not reported
“ware geometry conditioning and camera-guided noise further support per-view motion control and cross-view geometric exchange. We adapt the model for block-causal generation and distill the 30-step Teacher into a Student with four denoising steps. We train UMI-WM on self-collected open-world and in-studio UMI recordings combined with public manipulation data. To evaluate action following, we use an inverse dynamics mo”
Exact source · revision 1Source owner not reported
Reported quantity · teacher steps: 30 other · Basis not reported
“ware geometry conditioning and camera-guided noise further support per-view motion control and cross-view geometric exchange. We adapt the model for block-causal generation and distill the 30-step Teacher into a Student with four denoising steps. We train UMI-WM on self-collected open-world and in-studio UMI recordings combined with public manipulation data. To evaluate action following, we use an inverse dynamics mo”
Exact source · revision 1Source owner not reported
Reported quantity · reference-video fidelity metrics: 5 other · Basis not reported
“del (IDM) to recover relative camera poses and gripper widths from generated videos, and introduce Bimanual Action Fidelity (BAF) to measure how closely these recovered actions match the conditioning actions. UMI-WM achieves the best results on all five reference-video fidelity metrics and the best or competitive performance on VBench and WorldArena. The Teacher and Student outperform all compared baselines in action”
Exact source · revision 1Source owner not reported
Evidence dependency changes (2)
- 2026-10-30T22:20:36.684Z · evidence updated · source revision 1. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T22:20:20.649Z · evidence updated · source revision 1. Evidence extraction was updated; current source attributions are shown above.
Report timeline
Follow attributed reports and material updates.
- arXiv Robotics — research abstractsSignalUMI-WM: Interactive World Modeling for Biman-ual Manipulation from Wrist Cameras
UMI-WM is an action-conditioned bimanual world model that jointly generates both wrist views from observation history, per-arm relative camera poses, and gripper widths, with cross-view consistency, action following, and visual fidelity. The model outperforms baselines in action following and generates paired wrist videos 6.9 times faster than the Teacher model.
Timezone · UTC
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).