LinhanWangother
GlanceWAM reproduction bundle
A reproduction bundle for GlanceWAM (Sparse Test-Time Imagination for World-Action Models) containing pre-trained checkpoints, datasets, and code for evaluating world-action models on LIBERO and RoboCasa kitchen benchmarks.
Downloads12
Episodes4000
Likes1
Why This Matters for Physical AI
This dataset is critical for advancing world-action models by providing benchmark evaluations on real robot manipulation tasks with vision-language grounding, demonstrating how sparse test-time imagination can improve policy learning efficiency.
Technical Profile
- Modalities
- rgblanguage
- Robot Embodiments
- mobile_manipulator
- Action Space
- end_effector_delta
- Environment
- kitchensimulation
- Task Types
- manipulationpick_and_place
- Episodes
- 4000
- Data Format
- LeRobot
- Annotation Types
- language_instructionsaction_labels
- License
- other
Access
Need custom rgb data?
Claru builds purpose-built datasets for kitchen applications with dense human annotations and quality assurance.
Request a Sample Pack