6 papers
Silent Failures in Physical AI: A Literature Review of Runtime Action Authorization for Autonomous Systems
Barak Or
Physical AI systems increasingly map multimodal observations, language instructions, and learned world representations into physically consequential actions. Robotics foundation mo…
Can Predicted Dynamics Exist in the Physical World?
Barak Or
Can learned state-action proposals exist in the physical world? To filter infeasible commands before execution, policies are often wrapped in a runtime monitor. However, aggregatin…
Quantifying Prior Dominance in RAG Systems
Barak Or
Retrieval-Augmented Generation (RAG) grounds Large Language Models in external knowledge, yet current evaluations rely on discrete heuristics that suffer from ''epistemic blindness…
Kalman-Inspired Runtime Stability and Recovery in Hybrid Reasoning Systems
Barak Or
Hybrid reasoning systems that combine learned components with model-based inference are increasingly deployed in tool-augmented decision loops, yet their runtime behavior under par…
Automatic Stability and Recovery for Neural Network Training
Barak Or
Training modern neural networks is increasingly fragile, with rare but severe destabilizing updates often causing irreversible divergence or silent performance degradation. Existin…
Learning-Based Speed Estimation from Accelerometer-Only Inertial Sensing
Barak Or
The proposed model, CarSpeedNet, estimates scalar vehicle speed from a window of three-axis smartphone acceleration, without gyroscope, wheel-odometry, vehicle-bus, or positioning…