Seeing the Whole Picture: A Single Vector for Complete Scene Understanding
Researchers have developed a new framework that allows robots to grasp entire visual scenes from minimal information, paving the way for more robust and efficient learning.
Science
Why Explaining AI Redactions Builds Trust
New research reveals that transparency around how artificial intelligence obscures sensitive information is crucial for fostering user confidence in AI-driven communication.
Predictive Simulations: Stabilizing Robot Worlds with Reinforcement Learning
![A robot, guided by an action-conditioned world model, demonstrates a capacity for sustained, structurally-consistent video prediction-maintaining the integrity of a simulated object [latex] \text{over time} [/latex]-where competing methods rapidly succumb to accumulating error and object disintegration, establishing a new benchmark in predictive fidelity.](https://arxiv.org/html/2603.25685v1/x1.png)
Researchers are leveraging reinforcement learning to refine simulated environments, creating more reliable and consistent ‘world models’ for training and evaluating robotic systems.
The Hidden Order: How Machines Discover Physical Symmetry
![A symmetry-aware machine learning model’s predictive accuracy hinges on its adherence to group equivariance-specifically, whether its outputs transform predictably under symmetry operations-a condition quantified by metrics assessing both the variance of back-transformed predictions [latex]A_{\alpha}[/latex] and the decomposition of internal features [latex]B_{\alpha}[/latex] using Haar integration over the relevant symmetry group.](https://arxiv.org/html/2603.24638v1/x1.png)
New research reveals that machine learning models can independently learn fundamental physical symmetries, offering insights into their internal representations and the impact of neural network design.
AI Takes the Reactor: Automating Complex Nuclear Simulations
A new framework leverages artificial intelligence to drastically reduce the time and effort required to prepare input files for nuclear reactor modeling.
Harmonious Motion: Smarter Control for Collaborative Robots
New research introduces a control strategy that optimizes mobile manipulator movements during physical human-robot interaction, resulting in a more natural and safer experience.
Unlocking Molecular Motion with AI
![An explainable deep learning framework identifies reaction coordinates by mapping candidate collective variables to a neural network - consisting of [latex]N_{\mathrm{layer}}[/latex] layers and [latex]\mathbf{N}_{\mathrm{node}}[/latex] nodes - trained to approximate the sigmoidal function [latex]p_{\mathrm{B}}(q) = [1 + \tanh(q)]/2[/latex], thereby linking the reaction coordinate [latex]q[/latex] to the committor [latex]p_{\mathrm{B}}[/latex] and enabling analysis of free-energy landscapes.](https://arxiv.org/html/2603.25237v1/x1.png)
A new deep learning framework leverages explainable AI to reveal the key factors driving changes in complex molecular systems.
Chemistry’s AI Assistants: Automating Complex Simulations
Researchers are leveraging artificial intelligence to streamline and accelerate multistep computational chemistry workflows, moving beyond single-step predictions.
Coordinated Robot Teams Navigate Uncertainty with Dynamic Goals
A new framework empowers multi-robot systems to reliably coordinate in changing environments with moving targets, even when faced with unpredictable conditions.
Feeling Around in the Dark: Robots Learn to Handle the Unpredictable
New algorithms enable robots to safely explore and manipulate objects in uncertain, deformable environments without prior knowledge of their properties.