Safe Agent Execution Gets Localized Obligations
HALO proposes localized obligations for admitting and supervising heterogeneous agent actions safely.
Evidence Maturity: preprint
Temporal Constraints Target More Stable Latent Actions
The research introduces temporally constrained latent-action learning for more coherent embodied control.
A Humanoid Head Gains Personalized Episodic Memory
German researchers evaluated lightweight personalized memory for sustained human-humanoid interaction.
Cross-Embodiment Transfer Reaches Aerial Relay Teams
Korean researchers transfer structural motion priors from robot arms to coordinated aerial relay planning.
Action-Conditioned World Models Move Into Planning
Action-conditioned predictive models are used to improve generalizable embodied decision-making.
VLA Failures Become Corrective Training Signals
RedFlow converts failed robot actions into localized corrections for vision-language-action policy improvement.
World Models Receive a Security Threat Model
Researchers map cyber-physical attacks across sensing, world modelling, planning, control and human interaction.
Robots Are Asked to Reason About Their Own Aging
A framework integrates hardware degradation into autonomous planning, mission execution and maintenance decisions.
Teleoperation Data Gains an Action-Conditioned World Model
RynnWorld-Teleop combines action conditioning, depth awareness and skeletal structure for egocentric teleoperation prediction.
Embodied Operators Are Reframed as Deployable Systems
The work argues embodied operators should be evaluated as holistic deployable components rather than isolated models.
