Robot Foundation Models
PulseAugur coverage of Robot Foundation Models — every cluster mentioning Robot Foundation Models across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
RoboTryOn synthesizes robot demonstrations from human videos
Researchers have developed RoboTryOn, a novel system that synthesizes robot demonstrations from human videos using a virtual try-on approach. This method translates human movements into robot actions, preserving the ori…
-
ACCRUE framework accelerates robot learning, retaining skills across tasks · 2 sources tracked
Researchers have developed ACCRUE, a novel continual off-policy reinforcement learning framework designed to enhance robot foundation models (RFMs). ACCRUE enables robots to learn new tasks efficiently while retaining p…
-
New ARC method boosts robot foundation model performance without more data
Researchers have developed a new method called ARC, which enhances the zero-shot task performance of robot foundation models (RFMs) without requiring additional robot demonstrations or extensive training. ARC utilizes r…
-
New paper categorizes Generative Physical AI approaches for robotics
A new arXiv paper provides a comprehensive review of Generative Physical Artificial Intelligence (GPAI), a field that integrates large foundation models with physical robots. The paper categorizes GPAI systems into five…
-
New survey defines 'embodiment gap' in robot foundation models
A new survey paper published on arXiv introduces the concept of the "embodiment gap" in robot foundation models (RFMs). This gap refers to the difference between a reusable model and its practical deployment on a specif…
-
New framework steers robot models using human videos
Researchers have developed WAM-TTT, a novel framework designed to steer robot foundation models (RFMs) using human play videos. This method allows for adaptation without requiring additional robot demonstrations or task…
-
New SC3-Eval method uses video generation to evaluate robot foundation models
Researchers have developed SC3-Eval, a novel method for evaluating robot foundation models through self-consistent video generation. This approach addresses the limitations of real-world robot testing by simulating poli…