Ward: Education

109

Gaussian Pixel Codec Avatars: A Hybrid Representation for Efficient Rendering

arXiv:2512.15711v1 Announce Type: new
Abstract: We present Gaussian Pixel Codec Avatars (GPiCA), photorealistic head avatars that can be generated from multi-view images and efficiently rendered on mobile devices. GPiCA utilizes a unique hybrid representation that combines a triangle mesh and aniso…

222

USGS Develops Finer-Scale Approach to Find Areas of High Nitrogen Input to Groundwater

A recent study by the U.S. Geological Survey demonstrates that measuring nitrogen in stream reaches can identify areas where decreasing nitrogen inputs may improve the water quality of downstream receiving waters.

531

Interpretable neural networks help reveal the nature of dark matter

A research team from the Xinjiang Astronomical Observatory (XAO) of the Chinese Academy of Sciences has developed an interpretable artificial intelligence (AI) framework named Convolutional Kolmogorov–Arnold Network (CKAN), which sheds new light on the properties of dark matter at galaxy-cluster sca…

332

Curved representational Bregman divergences and their applications

arXiv:2504.05654v3 Announce Type: replace
Abstract: By analogy to the terminology of curved exponential families in statistics, we define curved Bregman divergences as Bregman divergences restricted to nonlinear parameter subspaces and sub-dimensional Bregman divergences when the restrictions are l…

319

Cross-Model Verification of Wall-Bounded Flows using Finite-JAX

arXiv:2509.25569v3 Announce Type: replace-cross
Abstract: Accurate prediction of wall-bounded flows remains central to advancing both theoretical understanding and computational methods in fluid mechanics. In this study, we perform a numerical simulation of channel flow using a complementary approa…

101

INFORM-CT: INtegrating LLMs and VLMs FOR Incidental Findings Management in Abdominal CT

arXiv:2512.14732v1 Announce Type: new
Abstract: Incidental findings in CT scans, though often benign, can have significant clinical implications and should be reported following established guidelines. Traditional manual inspection by radiologists is time-consuming and variable. This paper proposes…

109

Dense Associative Memories with Analog Circuits

arXiv:2512.15002v1 Announce Type: new
Abstract: The increasing computational demands of modern AI systems have exposed fundamental limitations of digital hardware, driving interest in alternative paradigms for efficient large-scale inference. Dense Associative Memory (DenseAM) is a family of models…

435

Roman soldiers defending Hadrian's Wall were infected by parasites, study finds

A new analysis of sewer drains from the Roman fort of Vindolanda, close to Hadrian's Wall, has shown that the occupants were infected by three types of intestinal parasite—roundworm, whipworm, and Giardia duodenalis.

111

Pioneers of 2D metals win the Physics World 2025 Breakthrough of the Year

Researchers describe their success with five metals as the “tip of the iceberg”
The post Pioneers of 2D metals win the Physics World 2025 Breakthrough of the Year appeared first on Physics World.

111

EPSM: A Novel Metric to Evaluate the Safety of Environmental Perception in Autonomous Driving

arXiv:2512.15195v1 Announce Type: new
Abstract: Extensive evaluation of perception systems is crucial for ensuring the safety of intelligent vehicles in complex driving scenarios. Conventional performance metrics such as precision, recall and the F1-score assess the overall detection accuracy, but …

111

Physicists Propose First-Ever Experiment To Manipulate Gravitational Waves

A new concept for energy transfer between gravitational waves and light. When massive cosmic objects such as black holes merge or neutron stars crash into one another, they can produce gravitational waves. These ripples move through the universe at the speed of light and create extremely small chang…

220

IMKD: Intensity-Aware Multi-Level Knowledge Distillation for Camera-Radar Fusion

arXiv:2512.15581v1 Announce Type: new
Abstract: High-performance Radar-Camera 3D object detection can be achieved by leveraging knowledge distillation without using LiDAR at inference time. However, existing distillation methods typically transfer modality-specific features directly to each sensor,…

633

Hybrid excitons: Combining the best of both worlds

Faster, more efficient, and more versatile—these are the expectations for the technology that will produce our energy and handle information in the future. But how can these expectations be met? A major breakthrough in physics has now been made by an international team of researchers from the Univer…

464

Hidden viruses: Amoebae as a water safety 'Trojan horse'

Human norovirus and adenovirus, two major causes of viral gastroenteritis, can persist for extended periods inside free-living amoebas that are common in natural and engineered water systems. A newly published study shows that these gut viruses can "hide" within different stages of the amoebas and r…

323

Private donors pledge $1 billion for world's largest particle accelerator

Europe's physics lab CERN on Thursday said private donors had pledged $1 billion toward the construction of a new particle accelerator that would be by far the world's biggest.

121

This 1.5-Million-Year-Old Human Face Is Rewriting Human Evolution

A newly reconstructed fossil face from Ethiopia reveals surprising complexity in early human evolution. By digitally fitting together teeth and fossilized bone fragments, researchers reconstructed a strikingly well preserved face of a human ancestor that lived 1.5 million years ago. The find represe…

220

Safety with Agency: Human-Centered Safety Filter with Application to AI-Assisted Motorsports

arXiv:2504.11717v4 Announce Type: replace
Abstract: We propose a human-centered safety filter (HCSF) for shared autonomy that significantly enhances system safety without compromising human agency. Our HCSF is built on a neural safety value function, which we first learn scalably through black-box …

220

Bridging the Gap Between Modern UX Design and Particle Accelerator Control Room Interfaces

arXiv:2512.14872v1 Announce Type: cross
Abstract: Accelerator control systems often represent relatively complex and safety-sensitive human-machine interfaces within process control industries. These systems are technically robust and reflect the cumulative integration of solutions built and adapte…

110

Vision-based module for accurately reading linear scales in a laboratory

arXiv:2512.15327v1 Announce Type: new
Abstract: Capabilities and the number of vision-based models are increasing rapidly. And these vision models are now able to do more tasks like object detection, image classification, instance segmentation etc. with great accuracy. But models which can take acc…

120

Improving Action Smoothness for a Cascaded Online Learning Flight Control System

arXiv:2507.04346v3 Announce Type: replace
Abstract: This paper aims to improve the action smoothness of a cascaded online learning flight control system. Although the cascaded structure is widely used in flight control design, its stability can be compromised by oscillatory control actions, which p…