Researchers introduce PACE, a framework that enables humanoid robots to dynamically generate and adapt psychologically grounded personas through conversational interaction with users. The system employs an interactive elicitation pipeline to create structured identities, which are then expressed via multimodal behaviors on the Ameca robot. Empirical evaluation demonstrates that dynamically generated personas improve user trust, perceived anthropomorphism, and interaction quality compared to static, generic personas.
Why it matters: This work represents a significant advance in human-robot interaction by enabling robots to personalize their identities in real time, which can enhance trust and engagement in practical applications.
MemoGuard is a lightweight adaptive runtime system that validates episodic memories against safety contracts before reuse in communication-limited robot navigation. In simulation, MemoGuard reduced battery safety violations by 76.6% compared to similarity-only memory reuse and decreased fallback calls by 21.4% relative to always invoking local reasoning. The system was evaluated on an NVIDIA Jetson AGX Xavier and is open-sourced.
Why it matters: MemoGuard advances the safety and efficiency of autonomous robots operating in mission-critical, communication-limited environments by reducing unsafe memory reuse and computational overhead.
SkillNav is a new behavioral skill framework for vision-language model (VLM)-based navigation that leverages curiosity value maps as writable memory, enabling composable skills without retraining. The method introduces a tiered skill intervention system and achieves state-of-the-art SPL on MP3D (25.5), HM3D v0.1 (39.3), and HM3D v0.2 (43.2), with improvements of up to 6.0 SPL over previous methods. SkillNav also attains the highest reported Success Rate on HM3D v0.1 (69.7) and v0.2 (75.9).
Why it matters: SkillNav enables continual, training-free improvement of navigation agents by allowing new skills to be added without retraining, advancing the flexibility and practicality of embodied AI systems.
Researchers have developed OptiTac, a biomimetic tactile sensor that uses optical fibers paired with mechanical pins to emulate the mechanoreceptor-to-nerve architecture of human touch. This design enables routing of tactile information away from the sensing surface while maintaining high spatial resolution. The system uses simple analytical methods, rather than deep learning, to infer contact properties such as location, size, and shape, resulting in interpretable tactile intelligence.
Why it matters: OptiTac offers a scalable and interpretable approach to distributed tactile sensing in robots, advancing the development of human-like touch perception.
Researchers introduce a fast-slow architecture for autonomous driving that separates a frozen 7B vision-language model (operating at 5 Hz) from a lightweight action expert (operating at 20 Hz). The slow model provides a cached scene representation, while the fast expert uses this cache and the current camera frame to generate waypoints at every simulation tick. In CARLA simulations on LangAuto-Short routes, this approach achieves 94% route completion compared to 37% for a frame-skipping baseline, and per-tick freshness reduces red-light violations by a third. The system also demonstrates strong zero-shot transfer to unseen towns and significantly lowers waypoint error compared to the backbone model's own action head.
Why it matters: This work offers a practical solution to bridging the latency gap between large language model inference and real-time control in autonomous driving, enabling substantial improvements in closed-loop performance and generalization.
Researchers introduce AC-VLA, a plug-and-play framework designed to improve out-of-distribution (OOD) generalization in Vision-Language-Action (VLA) models for robotics. The approach addresses two key failure modes—trajectory overfitting and perceptual shortcuts—by combining an LLM-driven instruction decomposer with a state-conditioned masking strategy. AC-VLA achieves approximately 28% absolute improvement on compositional OOD tasks in benchmark tests, while maintaining strong in-distribution performance.
Why it matters: Improving OOD generalization in VLA models is crucial for enabling robots to reliably perform novel task combinations in real-world environments.
AEGIS is a two-layer system for open-source liquid handlers such as the Opentrons OT-2. The first layer uses a large language model (LLM) combined with an assay rule database to detect protocol violations, achieving an F1 score of 0.97 on a multi-assay benchmark. The second layer applies PCA to pipette trajectories for runtime failure detection, reaching an average precision of 0.89. The system is open-source and demonstrates low-cost monitoring through cascade triage.
Why it matters: AEGIS enables reliable, low-cost detection of both protocol and runtime failures in open-source liquid handling robots, addressing a key gap in self-driving laboratory automation.
Researchers introduce a Generative Partially Specified Finite State Machine (GPSFSM) neurosymbolic architecture for robot behavior planning, combining the structure of FSMs with large language models. The system demonstrates higher plan-generation success rates than the state-of-the-art BTGenBot, especially in zero-shot scenarios, while maintaining comparable or lower planning latency. An open-source ROS2 stack implementing this approach is released.
Why it matters: This work presents the first generative FSM framework for robotics, offering a computationally efficient and interpretable alternative to existing behavior planning methods.
Researchers have introduced a comprehensive benchmark for the robotic arm reach-avoid task using deep reinforcement learning (DRL), employing MuJoCo MJX and Brax for parallelized simulation. The benchmark covers a diverse range of settings, achieving state-of-the-art success rates up to 98.8% for reach tasks and 95.2% for static reach-avoid tasks. However, the study finds that DRL agents' performance drops significantly in more realistic, complex scenarios, highlighting current limitations.
Why it matters: This work exposes the gap between DRL performance in simplified versus realistic robotic control tasks, emphasizing the need for further research to address real-world complexities.
Xiaomi has introduced Xiaomi-Robotics-1, a vision-language-action (VLA) model trained on more than 100,000 hours of real-world manipulation trajectories. The model demonstrates strong scaling behavior, achieving new state-of-the-art results on the RoboCasa365 benchmark (57.6% success rate) and RoboDojo (20.07 average score), surpassing previous bests. Xiaomi-Robotics-1 can also be efficiently fine-tuned for novel downstream tasks with minimal data.
Why it matters: This work establishes a new performance benchmark for robotics foundation models, highlighting the impact of large-scale real-world data and model scaling on generalization and adaptability.
ImprovedVBGS is a new framework for real-time, on-the-fly 3D reconstruction, designed for applications in robotics and autonomous navigation. It achieves a 1680x speed-up over previous Variational Bayes Gaussian Splatting (VBGS) methods, reducing per-frame latency from approximately 84 seconds to 0.050 seconds on an RTX 3070 Ti, while maintaining reconstruction quality. The acceleration is accomplished through spatially truncated variational inference and improved reassignment strategies.
Why it matters: This advance enables practical, real-time continual 3D reconstruction for robotics and autonomous systems operating under strict latency and memory constraints.
Nvidia is expanding its physical AI ecosystem with updates that include foundation models, edge hardware, software, developer tools, and industrial partnerships. The company is aiming to strengthen its presence in robotics and edge AI.
Why it matters: This move highlights Nvidia's efforts to build a comprehensive physical AI platform, which could accelerate the adoption of robotics and edge AI across industries.
Agility Robotics is opening a new training center for its Digit robots in Fremont, California. The facility will support the deployment and operation of the company's humanoid robots.
Why it matters: The new center places Agility Robotics in close proximity to other robotics and automotive companies, highlighting the growing competition in the humanoid robotics sector.
Researchers introduce FLARE, a physical spotlight attack that can reduce Vision-Language-Action (VLA) model task success rates to zero without requiring access to model internals. They find that standard adversarial training causes models to ignore color information, leading to degraded performance on color-dependent tasks. To address this, they propose ChromaGuard, a chroma-preserving defense that restores high success rates—97.5% on benign and 92.5% on attacked tasks—on a real robot platform.
Why it matters: This work exposes a critical vulnerability in VLA models to lighting-based attacks and demonstrates a practical defense, informing the deployment of robust robot manipulation systems in real-world environments.
MEMORA introduces Embodied Action Memory (EAM), a structured memory system for long-horizon robot planning that mimics human-like recall of past experiences. The system uses a formation-consolidation-retrieval lifecycle with four distinct memory stores and is evaluated on 45 hours of EPIC-KITCHENS-100 video. MEMORA demonstrates improvements in memory-assessment accuracy by up to 20.5 points and robot-grounded plan scores by 16.6% over baseline methods.
Why it matters: This work advances robot planning by enabling systems to leverage structured, persistent memories of past experiences, leading to more effective long-term reasoning and action.
CosFly-VLA is a spatially aware vision-language-action (VLA) model designed for UAV tracking in complex environments with frequent occlusions. The model jointly grounds targets, estimates their visibility, and generates flight actions using a structured prediction interface. Its multi-stage training pipeline includes spatial pretraining, curriculum-based fine-tuning, chain-of-thought reasoning, and reinforcement learning. Compared to OpenVLA, CosFly-VLA achieves over 34% reduction in displacement error and up to 29.8% improvement in success rate on benchmark tests.
Why it matters: This work demonstrates a significant advance in robust UAV tracking by integrating spatial reasoning and occlusion handling, enabling more reliable autonomous navigation in challenging real-world scenarios.
A new preprint proposes adopting the Agent-Client Protocol (ACP) as a unified communication standard for the human-robot interaction (HRI) layer in agent-based robotic systems. By combining ACP with the Model Context Protocol (MCP), the authors introduce a decoupled three-layer architecture that separates human interaction, deliberative orchestration, and physical execution. Experiments on a physical mobile robot demonstrate interoperability across three different user interfaces with negligible latency overhead.
Why it matters: Standardizing the HRI layer could reduce fragmentation and enable more flexible, real-time human-in-the-loop collaboration in robotic systems.
A new arXiv preprint presents a scaling recipe for Behavior Foundation Models (BFMs) in humanoid robot control. The authors coordinate motion tracking learning, on-policy rollouts with diverse reference motions, and a Humanoid Transformer architecture to improve control fidelity and generalization. Their approach achieves substantial reductions in Mean Per-Keypoint Position Error—over 10% in local mode and 82% in global mode—compared to existing humanoid controllers, as demonstrated in both simulation and real-world tests.
Why it matters: This work offers a principled and effective framework for scaling BFMs, advancing the development of more capable and general-purpose humanoid robots.
A new cloud-edge multimodal interaction framework for robots integrates an enhanced YOLO-based gesture detector with large language and vision-language model agents. The system achieves high precision (98.9% on a public dataset) and strong task success rates (95% for single-action tasks) while offloading intensive computation to the cloud. User evaluation with 30 participants reported a mean satisfaction score of 3.69 out of 5.
Why it matters: This work demonstrates a practical and effective approach to combining advanced gesture detection with large language and vision models for improving human-robot interaction in resource-constrained settings.
A new framework called ConFlow introduces constraint-guided flow matching for robot motion generation by integrating constraints directly into the training process using differentiable barrier or cost functions. The method leverages conditional Gaussian processes and uses infeasible demonstrations as negative supervision to enhance constraint satisfaction. Experiments on a two-robot navigation task show that ConFlow achieves lower collision rates and higher trajectory quality compared to standard flow matching approaches.
Why it matters: Integrating constraints during training addresses the gap between training and inference in generative motion models, potentially improving the safety and reliability of robot motion planning.