CoRL 2026 Call for Papers: Scaling Laws for Robot Learning from Human Data
mangahomanga · x · 2026-08-10
CoRL 2026 will host the Scaling H2R workshop, focusing on the central role of human data in scalable robot learning. The workshop explores how robot capability grows as a function of the volume, diversity, and embodiment gap of the human data collected.
Bringing together experts in imitation learning, egocentric perception, and hardware co-design, the event aims to address how to turn large, heterogeneous human data into reliable robot policies. Core topics include:
- Data Engines: Building human data engines and retargeting pipelines.
- Scaling Laws: Studying scaling laws and dataset composition.
- World Models: Developing egocentric world models from human video.
- Hardware Co-design: Co-designing hardware to close the human-robot gap.
The submission deadline is October 7, 2026.
Related event: CoRL 2026 to Explore Human Data Scaling Laws in Robotics(2 posts)→
More from Embodied
- Willing to Pay $20k-$30k Plus Software Costs for a全能 Home Robot — MickeySteamboat · 2026-08-10
- Humans temporarily win at parking; ex-Tesla engineers say human takeovers improve FSD — Scobleizer · 2026-08-10
- Humanoid Robots Will Gain Social Acceptance Faster Than Expected — VraserX · 2026-08-10
- CMU Lab Demonstrates Humanoid Robot Bending Banana Shots — TinfoilTricorn · 2026-08-10
- Nori Robotics Releases MotorLab: Open-Source Tool for Servo Management — Scobleizer · 2026-08-10
- Fun Demo: Hiring Miniature Optimus Bots to Build a Terafab Factory Entrance — bennash · 2026-08-10