New robotics and physical-AI papers, with what each one means for the industry.
Robotic foundation models offer a promising path toward general-purpose humanoid robot control, often through hierarchical architectures. However, their effectiveness depends on the command interface between the planner and the controller, which must support accurate execution while remaining easy to predict, and ideally allow new behaviors to be composed from prior ones.
Generative modeling is widely used for producing diverse objects from complex, multimodal distributions. However, its expressivity does not, in general, come with formal guarantees that the generated objects satisfy hard constraints or requirements.
Constrained Reinforcement Learning has recently gained increasing attention in the field of Safe Autonomous Driving, where the general mechanism is to maximize the expected reward while keeping the overall action risk bounded. In this way, the safety issues arising in AD can be mitigated through constrained actions.
Real-world robot learning is constrained by the cost of collecting experience and evaluating candidate behaviors. Video generation models offer a scalable foundation for visual simulators that predict action outcomes before physical execution.
We study robotic in-context learning (ICL), an emerging paradigm that enables robots to infer and execute tasks from visual demonstrations. Despite its growing promise, the problem itself remains under-defined: a visual demonstration simultaneously conveys action trajectories, object semantics, manipulation affordances, spatial relations, and task goals, making it unclear what information the robot is actually expected to follow.
This arXiv preprint introduces DEWO, a post-deployment learning method for world-action models that learns from observed visual futures around interaction turning points, not just action imitation. Across five simulated dexterous tasks, DEWO improves average success for multiple world-action model variants, and on two real dexterous-hand platforms it lifts success on initially viable spatial cells from 51.0% to 71.7% over two deployment rounds. The work matters because it shows continued predictive learning from deployment experience can make dexterous policies more reliable.
Robust and reliable perception is essential for autonomous robots operating in real-world environments, particularly in long-term missions where environmental conditions may change significantly over time. Although recent advances in Visual Foundation Models (VFMs) have improved open-vocabulary semantic segmentation, these models can still suffer from domain shift, which can significantly degrade performance if they are not adapted to the current environment.
Behavior Foundation Models (BFMs) give humanoids a promptable policy over a latent behavior space, enabling one single vector to represent a motion to imitate, a pose to reach, or a reward to maximize. Forward-Backward representations successfully produce such spaces, but at the cost of hundreds of GPU-hours for a single robot.
Encore is a preprint framework that distills a small number of teleoperated demonstrations into structured evidence for a coding agent, which writes, runs, and repairs manipulation policy programs without access to a runtime success signal. On 60 perturbed LIBERO-PRO cells, programs built from three demonstrations score 96.3% sealed success versus 94.9% without demonstrations, and on a real bimanual robot two tasks score 9/10 each. The result matters because demonstrations act as a search prior for auditable, frozen code policies rather than opaque imitation networks.
This arXiv preprint introduces self-compensating VLA, a deployment-time adaptation method that updates a robot policy online using the residual between commanded and executed motion, plus RoboStress, a simulation benchmark for joint-level execution errors. On two physical Piper arms, it improves average task success by more than 30 percentage points per arm, and on RoboStress it beats base policies and training-time robustness methods. The work matters because mechanical wear, payload, and thermal drift cause real execution errors that standard VLA training does not address.
Embodied systems must make knowledge acquired during one encounter usable in another despite changes in viewpoint, motion, and illumination. Yet aggregate cross-video accuracy conflates failures of local perception with failures to preserve observation identity, establish correspondence, and compose evidence, obscuring whether local video understanding actually transfers.
We propose a QCQP-representable IMU pre-integration factor that enables certifiable estimation with pre-integrated inertial measurements. To the best of our knowledge, this is the first work to directly incorporate IMU pre-integration into certifiable estimation.
While contact-rich manipulation requires deliberate regulation of interaction forces, recent approaches to robot manipulation learning predominantly represent actions as target positions or poses. Even methods that incorporate force sensing either use it solely as an observation or, when predicting forces as part of the output, rely on a hybrid force controller.
Large language models (LLMs) are increasingly used as high-level planners in robot navigation, but their outputs may become unreliable when instructions are ambiguous, unsupported by the environment, or semantically inconsistent. This paper presents a Risk-Aware Semantic Grounding framework for trustworthy LLM-based robot planning.
In this preprint, the authors build an onboard perception pipeline that lets a lightweight Explorer robot find minerals in underground mines from free-text prompts and hand compact, oriented 3D inspection targets to a heavier Inspector robot. Field tests in a test mine and an active magnesite mine detected two mineral types without retraining and condensed about 38,000 raw mineral points into a few dozen polygons, roughly two orders of magnitude smaller. This matters because autonomous mining inspection requires no mine-specific labels and low-bandwidth robot-to-robot handover.
Vision-language models can coordinate long-horizon robot manipulation, yet successful task reasoning still depends on whether local physical interactions produce the intended effects. We study how repeated interaction can improve this capability without updating the base model.
Grinding and sanding are fundamental processes in industrial robotic surface finishing. However, physical trials are expensive and consume workpieces, making reproducible experiments difficult.
Robots deployed in the physical world must be able to improve beyond their initial training as they encounter new situations and failures. For this improvement to scale across tasks, it must make effective use of experience without requiring human demonstration of each correction.
Diffusion and flow-matching Vision-Language-Action (VLA) policies generate action chunks through iterative denoising, incurring substantial inference latency that severely limits real-time robotic control. Existing acceleration methods treat an action chunk as a monolithic computational unit, ignoring a crucial physical reality of receding-horizon control: actions are generated jointly but consumed sequentially, resulting in inherently heterogeneous execution urgencies.
Video-based policy learning is particularly promising, as it illustrates target behaviors without requiring action annotations or embodiment-matched demonstrations. A central challenge is deciding what information should be transferred from the video to the robot.
Papers come from arXiv robotics feeds. Where our summary is not written yet, you see the opening lines of the abstract.