CVPR 2026 Workshop - World Models Meet Active Sensing and Closed-Loop Planning

the vision

Generative models have mastered passive generation. But real intelligence is active. It observes, plans, acts, and learns from feedback.

Active Sensing

Models that strategically choose what to observe—optimal viewpoints, sensor placement, information seeking.

Closed-Loop Planning

Continuous replanning based on new observations. Perception and action form a tight feedback loop.

Embodied Intelligence

Agents that learn through interaction. Active decision-making transforms passive models into interactive systems.

speakers

Leading voices in vision, robotics, and embodied AI

Yiannis Aloimonos

University of Maryland

Chelsea Finn

Stanford University

C. Karen Liu

Stanford University

Jitendra Malik

UC Berkeley

Nicholas Roy

MIT CSAIL

* listed alphabetically by last name

Call for paper

Submit 4-page short or 8-page long papers on world models, active sensing, embodied planning, and related topics. Non-archival, so work in progress is welcome.

Deadline: May 15, 2026
Notifications: May 22, 2026

submit a paper via OpenReview

sponsor

Lambda generously sponsors compute credits for outstanding research:

Best Paper Award — $3,000 in compute credits
Two Runner-Up Awards — $1,500 in compute credits each
Every Accepted Paper — $400 in compute credits

organizers

listed alphabetically by last name

Organizing & Submission Committee

half-day workshop • 8am–12pm

Location: Room 2A at conference center

Time: June 3rd, morning