1. RoboticsScore73

    SafeHarness: Enhancing Safety in Coding Agents for Robot Manipulation

    Researchers have developed SafeHarness, a system designed to improve the safety of coding agents used for robot manipulation. The system addresses the tendency of these agents to prioritize task completion over avoiding obstacles, a critical safety concern in real-world applications.

    Source: arXiv (2026-09-17) Full analysis
  2. Research PapersScore74

    Agile-WAM: An Agile Tactile World Action Model for Contact-Rich Robot Control

    Researchers have introduced Agile-WAM, an agile tactile World Action Model designed for contact-rich robot control. This model improves inference efficiency and deployment flexibility compared to previous tactile WAMs that relied on large generative backbones.

    Source: Agile-WAM: An Agile Tactile World Action Model for Contact-Rich Robot Control (arXiv preprint) Full analysis
  3. Research PapersScore74

    OPTED: On-Policy Fine-Tuning for End-to-End Driving using a Render-Free Teacher

    Researchers propose OPTED, a method for on-policy fine-tuning of end-to-end driving policies. OPTED decouples reinforcement learning from the fine-tuning process by using a privileged teacher model to supervise a pre-trained student model in closed-loop simulations.

    Source: OPTED: On-Policy Fine-Tuning for End-to-End Driving using a Render-Free Teacher (arXiv:2609.20756v1) Full analysis