The Intelligent Home: A New Framework for Secure, User-Centric Automation
Researchers unveil a novel blockchain-based system designed to bring greater security, efficiency, and user control to the rapidly evolving smart home landscape.
Science
Designing Photocatalysts with AI: Escaping the Limits of Trial and Error
![A computational workflow explores a design space of 820 candidates for COF photocatalysts, evaluating each through fragment-based screening-assessing band gap [latex] (IP−-EA) [/latex], conduction-band minimum, and stability-and comparing the efficacy of random sampling, Bayesian optimization, and an LLM agent that iteratively refines candidate selection over 200 iterations based on explicit chemical reasoning and quantitative feedback.](https://arxiv.org/html/2603.05188v1/2603.05188v1/fig_1_Ara.jpg)
Researchers have developed an artificial intelligence agent that dramatically accelerates the discovery of durable materials for capturing solar energy and driving chemical reactions.
Beyond Search: Can AI Truly Make Sense of the News?
A new benchmark challenges information-seeking agents to move beyond simple question answering and demonstrate genuine cross-document understanding on rapidly evolving, high-traffic topics.
Knowing When to Ask: Smarter Robots with Adaptive Decision-Making
A new framework allows vision-language-action models to dynamically adjust their approach-acting, reasoning, or deferring-based on perceived task difficulty.
Building Robots That Can Truly Help: Data’s Role in Adaptable Assistance
New research shows that strategically generated data can unlock open-set corrective assistance, enabling robots to handle unexpected challenges and new tasks in complex environments.
Guiding AI Through Proofs: A New Approach to Theorem Prediction
Researchers have developed a method that helps artificial intelligence navigate complex, multi-step mathematical proofs by leveraging the inherent structure of theorems themselves.
Smarter Robots: A New Approach to Motion Control
![The proposed control architecture unifies Model Predictive Control (MPC) with feedback mechanisms, establishing a robust and mathematically grounded system for dynamic regulation and optimization-a synthesis demonstrably superior to traditional, open-loop approaches given its ability to account for system uncertainties and disturbances through the incorporation of real-time measurements and corrective actions, formalized as [latex] u(k) = f(x(k), x_{ref}) [/latex], where [latex] u(k) [/latex] represents the control input at time step <i>k</i>, [latex] x(k) [/latex] is the system state, and [latex] x_{ref} [/latex] denotes the reference trajectory.](https://arxiv.org/html/2603.04988v1/2603.04988v1/x2.png)
Researchers have developed a unified control architecture that blends predictive modeling, feedback regulation, and machine learning to enable more precise and responsive movements in complex robotic systems.
Seeing is Sorting: AI-Powered Textile Recognition for Automation
A new approach leverages digital twins and advanced visual AI to dramatically improve the accuracy and efficiency of automated textile sorting systems.
Ask and Simulate: An AI Agent for Complex Physics
Researchers have developed an AI system that translates natural language requests into fully executable multiphysics simulations, lowering the barrier to advanced scientific computing.
Seeing, Understanding, and Acting: Guiding AI with Internal Signals
New research demonstrates a method for directly influencing the behavior of vision-language-action models by observing and manipulating their internal representations.