The paper introduces SafeHarness, an obstacle‑aware framework that improves the safety of coding agents for robot manipulation. By decomposing tasks into route planning and contact execution, the harness enables the agent to prioritize collision avoidance, achieving 71.9% task success and 87.5% collision avoidance—significantly better than prior methods. The study demonstrates that safety constraints can be effectively integrated into language‑model‑driven robot controllers.
arXiv:2512.00939v3 Announce Type: replace-cross Abstract: Recent progress in contact-rich robotic manipulation has been striking, yet most deployed systems remain confined to simple, scripted routine...
Physical Agentic AI proposes an architecture that links semantic planning with physical execution for robot crews. Each robot exposes a typed skill library, while a foundation model planner decomposes tasks into phases and assigns robot‑skill pairs. A Robot Orchestrator validates and authorizes one skill at a time, ensuring actions are grounded in robot capabilities, system state, and workflow constraints before actuation.
arXiv:2508. 19186v2 Announce Type: replace-cross Abstract: Reactive obstacle avoidance methods often cause agents to become trapped in local minima, because they can often only reason one step ahead (i.
arXiv:2606. 09749v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models have demonstrated impressive end-to-end performance across a variety of robotic manipulation tasks.
CrossSafe proposes embodiment-conditioned safety filtering that uses a Hamilton‑Jacobi reachability value function shared across robots while conditioning on each robot’s morphology and kinematics via a morphology‑aware latent representation. The method performs reachability analysis directly in latent space, enabling a single policy trained on multiple bimanual robot embodiments and manipulation tasks to generalize zero‑shot to a held‑out embodiment and reduce collision rates. Experiments on five embodiments and five tasks demonstrate that training with more embodiments improves generalization.
arXiv:2607. 11119v1 Announce Type: cross Abstract: Robot manipulation is a complex task that requires visual understanding, physical reasoning, planning, and closed-loop control.
The paper introduces URAI, a Universal Robot‑Agent Interface that separates robot control into two roles: a programming agent that writes reusable, task‑specific tools from intent, and an execution agent that calls these tools in a feedback loop. This design keeps high‑level decision making in the model while delegating low‑level motion to code, allowing tool revisions to persist across episodes without retraining the foundation model. Experiments on RoboDojo and AgileX tasks show significant gains in success rate, speed, and token efficiency compared to direct fingertip control and pre‑written programs.
arXiv:2608. 16889v1 Announce Type: cross Abstract: Long-horizon robot manipulation chains many contact-rich skills into one multi-stage task.
The paper introduces FailBank, a four‑stage self‑evolving framework that transforms runtime feedback from safety shields into lasting policy improvements for vision‑language‑action (VLA) models. By using a counterfactual correction teacher, outcome‑aware admission, and guarded LoRA updates, FailBank converts useful shield proposals into corrective targets while preserving successful actions as anchors. Experiments on the VLA‑Arena benchmark show that FailBank boosts task success rates by up to 8.5 percentage points and reduces cumulative policy cost by up to 35.6%, outperforming both base policies and traditional runtime shielding.
The paper introduces Collective Counterfactual Planning (CCP), a formal model describing how teams coordinate tasks that no single member can handle alone, constrained not by capability but by representational geometry. CCP defines four critical gates—exogenous implementation coalitions, conception, consent, and task-relative verification—that determine whether a team can achieve and legitimately recognize a conjunctive goal. The authors present the Collective Counterfactual Solvability (CCS) problem, separating geometric feasibility, executable attainment, and validated completion, and provide a sound and complete four-step solvability scheme under exact representation of relay closure.
arXiv:2606. 05805v1 Announce Type: new Abstract: LLM-based guardrails typically safeguard agents by evaluating proposed actions or inputs before execution, producing safety signals such as binary allow/deny decisions, risk categories, and/or explanatory rationales about potential policy violations.
The paper introduces SCOPE, a method that post‑trains computer‑use agents to balance task completion with safety by conditioning actions on environmental risk. It combines supervised fine‑tuning on three trajectory types—capability demonstrations, safe continuations, and explicit refusals—followed by reinforcement learning to improve performance. Experiments starting from Qwen3.5‑9B show that SCOPE‑RL achieves high task success and attack‑avoidance rates, outperforming other agents on OSWorld and OS‑BLIND benchmarks.