Mapping the Unknown: Robotic Teams Navigate Communication Limits
A new framework empowers fleets of robots to simultaneously explore and inspect large, complex 3D environments even with restricted communication.
Science
Teaching Robots to Understand Praise
New research introduces a dataset and benchmark for training robots to interpret natural language feedback, paving the way for more intuitive human-robot interaction.
Droplets That ‘See’ With Chemistry
Researchers have engineered microswimmers capable of autonomously navigating complex environments by sensing the chemical traces of their own movements.
Scaling AI with Light: A New Era of Photonics-Powered Computing
Researchers are charting a course toward large-scale artificial intelligence systems powered by photonics, demanding a complete overhaul of design methodologies.
Robots Get Their Bearings: AI-Powered Exploration and Object Retrieval
A new system combines large language models with robotic platforms, allowing robots to understand instructions, map environments, and autonomously locate and grasp objects.
Beyond Lists: Smarter Recommendations with Generative Slate Planning
A new framework, HiGR, leverages advanced generative techniques to move beyond simple item lists and create more diverse and satisfying recommendation slates for users.
Bridging the Gap: AI Vision Guides Robots to Seamless Action
Researchers have developed a new framework that allows robots to react to visual instructions in real-time, achieving smoother and more efficient manipulation.
Beyond Algorithms: Reimagining Recommendations with Large Language Models
A new framework, OpenOneRec, is leveraging the power of generative AI to build more flexible and performant recommendation systems.
Seeing the Parts: How Vision Transformers Build Images From Primitives
![The research details a compositionality framework applied to Vision Transformers, focusing on learning a composition function during Level 1 Discrete Wavelet Transform decomposition-specifically utilizing coefficients [latex]DaD_a[/latex], [latex]DbD_b[/latex], [latex]DcD_c[/latex], and [latex]DdD_d[/latex]-to enable a more nuanced understanding of feature representation.](https://arxiv.org/html/2512.24438v1/x3.png)
New research reveals that Vision Transformers decompose images into fundamental components, mirroring how humans understand visual scenes.
AI Agents Design Next-Gen Chips in Record Time
A new framework harnesses the power of artificial intelligence to automate semiconductor device design and optimization, dramatically accelerating the path to future technologies.