RLDX-1 Technical Report

While Vision-Language-Action models (VLAs) have shown remarkable progress toward human-like generalist robotic policies through the versatile intelligence (i.e. broad scene understanding and language-conditioned generalization) inherited from pre-trained Vision-Language Models, they still struggle with complex real-world tasks requiring broader functional capabilities (e.g. motion awareness, memory-aware decision making, and physical sensing). To address this, we introduce RLDX-1, a general-purpose robotic policy for dexterous manipulation built on the Multi-Stream Action Transformer (MSAT), an architecture that unifies these capabilities by integrating heterogeneous modalities through modality-specific streams with cross-modal joint self-attention. RLDX-1 further combines this architecture with system-level design choices, including synthesizing training data for rare manipulation scenarios, learning procedures specialized for human-like manipulation, and inference optimizations for real-time deployment. Through empirical evaluation, we show that RLDX-1 consistently outperforms recent frontier VLAs (e.g. $π_{0.5}$ and GR00T N1.6) across both simulation benchmarks and real-world tasks that require broad functional capabilities beyond general versatility. In particular, RLDX-1 shows superiority in ALLEX humanoid tasks by achieving success rates of 86.8% while $π_{0.5}$ and GR00T N1.6 achieve around 40%, highlighting the ability of RLDX-1 to control a high-DoF humanoid robot under diverse functional demands. Together, these results position RLDX-1 as a promising step toward reliable VLAs for complex, contact-rich, and dynamic real-world dexterous manipulation.

MA3DSG: Multi-Agent 3D Scene Graph Generation for Large-Scale Indoor Environments

Current 3D scene graph generation (3DSGG) approaches heavily rely on a single-agent assumption and small-scale environments, exhibiting limited scalability to real-world scenarios. In this work, we introduce Multi-Agent 3D Scene Graph Generation (MA3DSG) model, the first framework designed to tackle this scalability challenge using multiple agents. We develop a training-free graph alignment algorithm that efficiently merges partial query graphs from individual agents into a unified global scene graph. Leveraging extensive analysis and empirical insights, our approach enables conventional single-agent systems to operate collaboratively without requiring any learnable parameters. To rigorously evaluate 3DSGG performance, we propose MA3DSG-Bench-a benchmark that supports diverse agent configurations, domain sizes, and environmental conditions-providing a more general and extensible evaluation framework. This work lays a solid foundation for scalable, multi-agent 3DSGG research.

Towards Generalizable Scene Change Detection

Scene Change Detection (SCD) is vital for applications such as visual surveillance and mobile robotics. However, current SCD methods exhibit a bias to the temporal order of training datasets and limited performance on unseen domains; coventional SCD benchmarks are not able to evaluate generalization or temporal consistency. To tackle these limitations, we introduce a Generalizable Scene Change Detection Framework (GeSCF) in this work. The proposed GeSCF leverages localized semantics of a foundation model without any re-training or fine-tuning -- for generalization over unseen domains. Specifically, we design an adaptive thresholding of the similarity distribution derived from facets of the pre-trained foundation model to generate initial pseudo-change mask. We further utilize Segment Anything Model's (SAM) class-agnostic masks to refine pseudo-masks. Moreover, our proposed framework maintains commutative operations in all settings to ensure complete temporal consistency. Finally, we define new metrics, evaluation dataset, and evaluation protocol for Generalizable Scene Change Detection (GeSCD). Extensive experiments demonstrate that GeSCF excels across diverse and challenging environments -- establishing a new benchmark for SCD performance.

Manipulator control of the Robotized TMS System with Incurved TMS Coil Case

Objective: This study shows the force/torque control strategy for the robotized TMS system whose TMS coil's floor is incurved. The strategy considered the adhesion and friction between the coil and the subject's head. Methods: Hybrid position/force control and proportional torque were used for the strategy. The force magnitude applied for the force control was scheduled by the error between the coil's current position and the target point. Results: The larger desired force for the force controller makes the error quickly. By scheduling the force magnitude applied for the force control, the low error between the coil's current and target positions is maintained with the relatively small force after the larger force is applied for around 10 seconds. The proportional torque made the adhesion better by locating the contact area between the coil and the head close to the coil. I was shown by checking the ${\tau}_c/F_c$ value from the experimental results. While the head slowly moved away from the coil during the TMS treatment, the coil still interacted with the head. Using that characteristic, the coil could locate the new target point using the force/torque strategy without any trajectory planning. Conclusion: The proposed force/torque controller enhanced the adhesion between the incurved TMS coil and the subject's head. It also reduced the error quickly by scheduling the magnitude of the force applied. Significance: This study proposes the robotized TMS system's force/torque control strategy considering the physical characteristics from the contact between the incurved TMS coil case and the subject's head.