The authors collected 40,543 bimanual manipulation episodes spanning 1,718 hours and 151 tasks.
Dense subtask labels improve robots’ long-horizon bimanual manipulation
FineART-VLA learns to predict intermediate steps, improving spatial grounding, task completion and transfer between robot platforms.
AI Startup
Jade Choghari · Pepijn Kooijmans · Mansi Agarwal · Yusuf Umut Ciftci · Aseem Doriwala · Catherine Weaver · +5 more
Scale AI · Hugging Face · University of Southern California · Stanford University
Research Digest··2 min read
Choghari et al.
Why this paper
From Hugging Face and 3 others
In one line
Dense subtask supervision enables a bimanual vision-language-action policy to execute longer tasks, resolve spatial ambiguity, and transfer across robot platforms with less new data.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ·No stated limitations found
- ✓Reports numbers on named benchmarks (5 benchmarks)
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§