TraceVLA visual trace prompting dataset

From the catalogue. Listed by one researcher from one source page.

• id: tracevla-visual-trace-prompting-dataset
• kind: language and reasoning annotation
• robot or device: robot manipulation trajectories; evaluated on WidowX and SimplerEnv (class: cross embodiment)
• how collected: visual trace overlays on robot trajectories
• size as stated: 150K robot manipulation trajectories
• hours: not stated | hours only claimed: none | episodes: 150,000
• year: 2024
• organisation: umd-huang-lab (GitHub org)
• licence: not stated (class: not stated)
• access: not stated (class: not stated)
• link (dataset): https://github.com/umd-huang-lab/tracevla
• note: Whether the 150K set is downloadable was not confirmed.

Confirm the licence at the link before relying on it.