Scaling H2R workshop at CoRL 2026 probes scaling laws for human-to-robot learning
mangahomanga · x · 2026-11-12
Scaling H2R, a half-day workshop at CoRL 2026 (Nov 12, Austin, TX), shifts the human-to-robot question from feasibility to scale: robot capability as a function of how much and how diverse human data is collected.
Key points:
- Central question: do policy returns keep accumulating along hours, demonstrators, scenes, and embodiments, or saturate — and how does the human-robot embodiment gap affect scaling?
- Topics: human data engines and retargeting, scaling laws and dataset composition, egocentric world models from human video, hardware co-design.
- Invited speakers: Ruijie Zheng (NVIDIA), Shuran Song (Stanford), Homanga Bharadhwaj (JHU), Chelsea Finn (Stanford).
- Call for papers: 8-page long / 4-page short, non-archival, concurrent submissions welcome; deadline October 7, 2026 (AoE). Sponsored by NVIDIA and Tangible Robots.
More from Embodied
- CoRL 2026 workshop makes the case for continually self-improving robots — chris_j_paxton · 2026-11-12
- RobotEra humanoids sort parcels at SF Express hub during Double 11, 1,200 parcels per hour — CyberRobooo · 2026-11-12
- Hackathon Project Uses Hermes Agent to Turn Prompts into 3D-Printable CAD Designs — alexcovo_eth · 2026-11-12
- Tesla Robotaxi goes 24/7 across every city it operates in — EricETesla · 2026-11-12
- Direct-drive liquid-cooled robot hand debuts, targeting dexterity at the actuator level — ybisk · 2026-11-12
- CoRL attendees organize mechanical bull riding night to close the conference Texas-style — KyleMorgenstein · 2026-11-12