Robotics foundation models (RFMs) are making general-purpose robots increasingly practical for factory deployments. While RFM serving systems are central to this vision, existing systems are largely shaped by a single-robot, single-model assumption: inference is treated as an edge-computing problem handled by an on-robot or dedicated nearby GPU, and the serving objective is to minimize the latency of a single action model. In this paper, we propose ROSA, an RFM serving system for robot factories designed around three key principles.

Multimodal large language models (MLLMs) excel at visual interpretation but fail on spatial reasoning tasks that humans solve reliably. Existing benchmarks evaluate these models as black boxes, limiting their ability to identify the underlying causes of lower performance: when a model fails a spatial reasoning task, it remains difficult to ascertain whether the hurdle is perceptual, such as recognizing object boundaries, or cognitive, such as reasoning about occlusion to infer hidden geometry.

Gaussian-based image representations effectively model image content using compact parametric primitives while preserving high visual fidelity, yet storing a large number of floating-point parameters per primitive degrades rate-distortion efficiency at higher fidelity targets. To improve the rate-distortion performance in Gaussian representation, we present our Cluster-Guided Vector Quantization (CGVQ), a Gaussian primitive based image compression method.

Esports is having a significant societal impact. Yet scheduling competitive group play during experiments is difficult, and recreating competitive environments in the lab is challenging. To address these problems, we adopt the field experiment methodology, to create hybrid tournament-experiments — experimentation integrated into the tournament itself. We run two Rocket League tournament-experiments and discuss their methodological and experimental implications.

Non-verbal communication plays a critical role in multiplayer games, players often rely on gestures, movement patterns, item interactions, and UI signals to communicate intent, negotiate cooperation willingness, and avoid conflict. May et al. [May et al. 2013] demonstrated how non-verbal game behaviors are effective at changing player behavior and task success.

Large language model (LLM)-driven evolutionary search is an emerging algorithm-discovery paradigm that has already produced novel results in several scientific fields. Yet its application to wireless communications remains largely unexplored. To bridge this gap, we introduce The AI Telco Engineer (AITE), a framework to autonomously design algorithms for complex communication problems, while navigating performance-complexity tradeoffs.

Generative AI is rapidly transforming game development, enabling new approaches to content creation, simulation, and player interaction. However, for interactive systems, the central challenge is not generation alone, but control—the ability to steer generative models in real time, align them with player intent, and integrate them into production pipelines.

Esports is having a significant societal impact. Yet scheduling competitive group play during experiments is difficult, and recreating competitive environments in the lab is challenging. To address these problems, we adopt the field experiment methodology, to create hybrid tournament-experiments — experimentation integrated into the tournament itself. We run two Rocket League tournament-experiments and discuss their methodological and experimental implications.