policy

RoboScribe

omarwalid96 · PyTorch

or hover any field below to flag it

Overview

Name
RoboScribe
Author
omarwalid96
Framework
PyTorch
License
unknown
Skill type
other
Evidence level
untested
Task description
RoboScribe validates pre-trained robot motion policies using natural language commands — comparing RL policy behavior against VLM-generated actions to catch failures in simulation, while automatically generating training datasets for next-generation VLA models

Spaces

Action space
other · 0-dim · 0Hz
Observation space
  • type: other

Links

HuggingFace repo
null
Paper (arXiv)
null

Compatible robots

0+19 mentioned but not in catalog yet

No robots list RoboScribe as compatible yet. Know of one? Flag it above.

Compatible environments

0

No environments list RoboScribe yet.

Datasets that reference this policy

0

No datasets reference RoboScribe yet.