AI Takes the Helm: Mastering Nuclear Reactor Control
![An integrated framework for agentic physical AI demonstrates that scaling foundation models from 1K to 100K nuclear reactor scenarios induces qualitative phase transitions-increasing precision from 26.2% to 92%, collapsing variance by 500×, and compressing policy entropy from 1.38 to 0.89 nats-while a two-phase curriculum leveraging CPT and LoRA stabilizes agentic policies by separating domain structure from task specialization and concentrating 76% of actions on single strategies despite limited training frequency, ultimately achieving closed-loop control within specified tolerances in a physics-constrained environment [latex]\mathcal{M}\_{\text{feas}}[/latex].](https://arxiv.org/html/2512.23292v1/x1.png)
A new approach to artificial intelligence demonstrates reliable power control of a nuclear reactor through data-driven learning and closed-loop simulation.
Science
Guiding the Analyst: A Framework for Intelligent Assistants
New research maps the design landscape for AI agents that collaborate with humans in visual analytics, offering a systematic approach to building more effective data exploration tools.
Quantum Leaps in AI: A New Framework for Intelligent Systems
![The Schrödinger AI Energy Seeker demonstrates superior path planning capabilities compared to classical LSTM approaches, achieving efficient goal attainment through nuanced energy awareness-a characteristic that facilitates a natural trajectory unlike the habitual, and often inefficient, patterns exhibited by conventional models, as evidenced by its ability to optimize for [latex]E = \frac{1}{2}mv^2[/latex] during route calculation.](https://arxiv.org/html/2512.22774v1/replanning_figure.png)
Researchers are exploring the potential of quantum mechanics to build AI systems capable of more robust reasoning and generalization.
Dancing Robots: Controlling Humanoids with the Power of Sound
Researchers have developed a new method for controlling humanoid robots using audio signals, enabling more natural and expressive movement.
Streamlining Science: The AiiDAlab Approach
A new platform is emerging to tackle the challenges of complex computational workflows and accelerate discovery across scientific disciplines.
Wearable Futures: When Robotics Meets Runway
This review examines the emerging field of soft robotic fashion, exploring its potential to redefine personal expression and social interaction.
Beyond the Stars: The Rise of Intelligent Space Systems
A new era of autonomous spacecraft and data analysis is transforming both space exploration and life on Earth.
Bridging Brain and Machine: The Future of Memory
This review explores the convergence of cognitive neuroscience and artificial intelligence, examining how understanding human memory can inspire more effective systems for autonomous agents.
Learning from Watching: New Method Clones Behavior Directly from Video
Researchers have developed a novel approach that allows machines to learn complex tasks simply by observing video footage, eliminating the need for labeled actions or reward signals.
Automating Chip I/O Design with AI
Researchers have developed a new AI-powered system that significantly accelerates the design of analog and mixed-signal integrated circuit input/output rings.