Computer Modeling

Algorithm-designed photonic circuits beyond human intuition

A team of researchers at Harvard and Max Planck Institute have developed three new functional components for photonic microchips using an inverse design algorithm. The compact designs are about 500 times smaller than conventional designs and offer a path toward higher-performance integrated light technologies.

Two Lehigh University AI projects selected for DOE Genesis Mission funding

Two Lehigh University AI projects have been selected for funding from the Department of Energy's Genesis Mission. The RIVER-AI project will improve flood- and water-level prediction, while the REACT project aims to accelerate reactor-scale fusion energy by developing an AI-enabled digital twin. These awards strengthen Lehigh University...

New computer graphics research: From volcanic lava to robot vacuum room exploration

Researchers at ISTA develop two modeling approaches to represent extremely deformable surfaces more realistically while cutting computational costs. They take inspiration from natural systems like volcanic lava and cake batter to create a wave-based approach, which tackles long-standing graphics problems from a new angle.

KAIST opens a new era of Webtoons: From “viewing” to “experiencing”

A KAIST research team developed an XR comics platform, ComiXR, that enables users to read and create comics in immersive environments. The platform demonstrated increased immersion when comic elements were positioned at different depths and incorporated sensory experiences such as facial expression tracking.

First-of-its-kind computer model of bacterial biofilms could support antibiotic resistance research

A new 3D computer model developed by the University of Surrey has shown how Pseudomonas aeruginosa grows and spreads its protective layer under constant fluid flow. The model's accuracy was validated through laboratory experiments, demonstrating potential for faster and smarter ways to understand bacterial behavior.

New plug-and-play software simplifies engineering design

Researchers at UCF College of Engineering and Computer Science developed CRAFTS, a plug-and-play software that simplifies engineering design. The software offers a user-friendly interface and modularity, making it accessible to users without extensive coding knowledge.

Uncanny valley effect observed in macaques through 3D animated monkey avatars

A new tool generates realistic monkey body animations, revealing an uncanny valley effect where macaques prefer less realistic avatars to highly realistic ones. The study provides the first evidence that non-human primates experience this phenomenon.

The real Moana story: Why the Polynesians suddenly sailed east

A team of geographers and climate scientists found that major drought in Samoa and Tonga forced Polynesians to migrate east into the Pacific. Climate modeling revealed increased rainfall in receiving islands and a shift in sea surface temperatures, creating powerful incentives for people to seek new opportunities.

Curtin partners with Octave to train next generation of digital builders

The partnership aims to bridge the gap between university learning and real-world practice, equipping students with in-demand digital skills. Through the Octave NextGen Builders Program, students will gain practical experience using professional tools used across industry projects.

AI mistakes can cost doctors time when writing to patients

A large-scale study of an online patient portal shows that AI-generated responses can introduce errors and extraneous details, leading to increased editing time for physicians. Adapting AI to individual physician communication styles can improve accuracy by 33% and reduce editing by 26%.

Building digital twins as decision infrastructure for a complex world

Digital twins are expanding rapidly, using real-time data to simulate and analyze systems before applying them in the real world. The study emphasizes the need for interoperable architectures, machine-readable metadata, and standardized trust frameworks to address challenges such as privacy, cybersecurity, and uncertainty.

UT climate model of last year’s July 4 storms suggests that sea surface temperatures actually reduced rainfall

Research Scientists Edward Vizy and Professor Kerry Cook analyzed a storm that caused catastrophic flooding in Central Texas. They found that warmer-than-average sea surface temperatures weakened the Great Plains low-level jet, resulting in weaker storms and less intense rainfall.

Computer model could enable bridges and buildings that use less material

MIT researchers developed a framework that allows users to apply constraints to algorithmically generated structures, making them more buildable. The approach has the potential to reduce carbon emissions in construction by up to 90% by designing structures with multiple materials and taking into account materials' properties.

On the trail of the missing hydrogen atoms

A scientific team has developed an AI-powered approach to solve a practical problem in materials science by locating missing atomic positions in otherwise known structures. By using an adapted open-source model called XtalPaint, researchers can reconstruct accurate crystal representations with a success rate of 97%.

AI fast-forwards molecular simulations by 10,000-fold

A new AI model has revolutionized molecular simulations, enabling researchers to predict molecular behavior and identify promising drug candidates more quickly. By analyzing over 12,500 organic molecules, the model has demonstrated accuracy and consistency with physical laws.

Old Dominion University launches National Security Institute to advance solutions for complex global challenges

Old Dominion University has launched a National Security Institute to accelerate research and technology solutions for pressing national security challenges. The institute will bring together researchers' strengths in AI, autonomy, and coastal systems to deliver innovative solutions.

Time for an AI checkup: Flaw found in machine learning for sepsis treatment

Researchers at Emory University discovered a flaw in reinforcement learning used to guide sepsis treatment, which can result in either overtreatment or undertreatment. The team developed a simple workaround to avoid the flaw, leading to an 8-10% decrease in patient mortality.

UNM researchers use new machine learning method to detect self-harm history hidden in veterans’ medical records

Researchers found that diagnosis codes captured only about one-fourth of clinically documented self-harm history, highlighting a common gap in health systems. A new study used a novel machine learning method, PULSNAR, to estimate the probability of self-harm history present but missing from diagnosis codes.

MIT researchers teach AI models to interpret charts

Researchers from MIT and IBM create a state-of-the-art dataset called ChartNet, which includes over a million varied charts. The dataset is designed to teach vision-language models how to effectively interpret charts, enabling them to outperform commercial models on tasks like data extraction and chart summarization.

Physics-trained digital ‘super-brain’ speeds up technology development

A digital 'super-brain' with physics-based knowledge significantly speeds up the design and development of optical components, such as those for quantum computers and camera lenses. By integrating physical principles into machine learning algorithms, researchers reduce simulation time from months to days.

New low-cost tool reveals hidden molecular switch points under light

A new quantum chemistry method predicts the behavior of molecules under light with lower computational cost, enabling the study of larger systems and complex reaction pathways. This breakthrough advances the discovery of next-generation materials and deepens understanding of molecular behavior under light.

When ‘sloppy’ decisions are actually smart

Researchers found that people who are less precise in their decision-making can outperform those who are more rational in certain social situations. This is because 'sloppy' decisions can help individuals avoid the pitfalls of highly sensitive learners, who prioritize short-term gains over long-term benefits. In contrast, in other situ...

Your own personal Farmville: This VR greenhouse lets users monitor crops remotely

Researchers at Binghamton University developed a system that enables users to monitor real plants in real-time using virtual reality. This technology makes farming more accessible for older adults and people with disabilities, allowing them to observe plants without physically being present.

Frontiers in Science Deep Dive webinar series | Soil science: how AI could help scientists secure a vital global resource

A new Frontiers in Science article explores how AI can accelerate scientific discovery in soil science by creating digital soil twins and trialing climate adaptation strategies. Researchers will discuss the potential of multi-agent AI systems to enable autonomous hypothesis generation, experimental design and data analysis during a fre...

Soil science: How AI could help scientists secure a vital global resource

A new study highlights the potential of AI tools in soil science, enabling researchers to better understand soil ecosystems and adapt to climate change. The system successfully generated hypotheses on how soils store carbon and what controls their storage limits, with outputs aligning with expert research.

Turning surroundings into a 'virtual screen' could help machines to better see in 3D

A new approach enables computers and machines to capture images at higher resolution and faster speed, making it impervious to reflective surfaces. The technology uses a virtual screen created by repurposing the surroundings of specular objects.

AI system automates coding for scientific research

A new AI system, Empirical Research Assistance (ERA), can automatically write scientific software programs that outperform human-written ones. ERA combines a large language model with search strategies to explore and refine thousands of pieces of code, reducing the time required for exploration from months to hours or days.

Technology that breaks down barriers in parasports wins Best Paper Award at CHI

A Harvard-led study developed BRIDGE, a simulation technology that reinterprets traditional non-disabled basketball footage into realistic wheelchair basketball video representations. The system significantly improved how natural player postures appeared and made tactical intentions easier to understand.

Study suggests data centers will boost power bills up to 57% by 2030

New research from North Carolina State University predicts data center power demand will increase by up to 57% by 2030, leading to higher electricity costs and CO2 emissions. The study finds that regional price increases will vary depending on where new data centers are built.

New scenarios needed to address climate crisis

Mathematical analysis reveals a hidden “golden rule” in abstract art

Researchers applied topological methods to distinguish between genuine and fake abstract artworks. They found that the structure of the compositions corresponds with how people perceive them. The study proposes an intuitive mathematical 'golden rule' for structuring art.

Researchers “reprogram” materials by quickly rearranging their atoms

New method to carry out long term climate simulations at fraction of cost

Aston University and STFC Hartree Center announce strategic partnership to advance neuromorphic computing in the UK

UC San Diego researchers launch free ‘digital twin’ for end-to-end testing of applications over wireless networks

AI models comb patient data to predict cardiac-arrest risk

Generative artificial intelligence can significantly reduce the number of animal experiments

Researchers developed genESOM, a generative AI that can expand dataset volume and simulate larger animal numbers while maintaining reliability. This allows for 30-50% reduction in animal experiments without compromising results.

New AI tool developed by Stowers Institute and Helmholtz Munich scientists predicts how cells choose their future — helping uncover hidden drivers of development

Researchers developed RegVelo, an AI framework that models cellular dynamics and gene regulation to predict cellular fate decisions. The model traces developmental trajectories and simulates regulatory interactions, providing insights into hidden drivers of development and potential therapeutic targets.

New study: Women are 60 percent more likely to be injured in car accidents than men

Deforestation lowers threshold for Amazon degradation to below 2°C warming

A new study reveals that around two-thirds of the Amazon rainforest could shift into degraded ecosystems at 1.5-1.9°C of global warming due to increased deforestation. The study highlights the critical role of the Amazon in regulating moisture recycling and stabilizing the Earth system.

A toy model to understand how AI learns

Researchers have developed a simplified mathematical model of learning in neural networks, shedding new light on how these systems produce their responses. The toy model, inspired by physics principles, captures key features of complex systems and offers insights into the surprising efficiency and stability of modern AI systems.

No digital content is safe from generative AI, researchers say

Researchers discovered that simple artificial intelligence tools can bypass security techniques meant to protect authentic content from use in deepfakes and facial identity theft. The study found that attackers can easily defeat existing image protection using off-the-shelf AI models and simple commands.

MIT engineers’ virtual violin produces realistic sounds

The new computational violin simulates the physics of string interaction with air, producing realistic sound. Luthiers can tweak parameters like wood type or body thickness before hearing the instrument's response.

New study shows antibodies need a strong core — not just grip — to fight SARS-CoV-2

Researchers identify mechanical stability as a crucial factor in antibody effectiveness against SARS-CoV-2. The study found that conventional antibodies distribute mechanical load unevenly and that viral mutations influence both binding affinity and mechanical resistance.

How a free flow of information can amplify incorrect ideas

Simulations by a team of scientists show that unrestricted information sharing can lead to less accurate collective beliefs, particularly in socially homogenous settings. This phenomenon is driven by homophily and the amplification of erroneous ideas within groups.

Simple robots that collectively build and excavate are inspired by ants

Researchers developed a fleet of cooperative robots that, like ants, can spontaneously organize to build and dismantle structures without central control. They identified key parameters needed for excavation and building performance using simple local rules.

A tool based on cell phone calls helps plan cities based on their residents' routines

Researchers at the University of Cordoba developed MAPLID, a tool that maps city structures and pace based on how residents use their cell phones. The multi-label classification provides information on population density changes and traffic patterns.

Artificial Intelligence in Nephrology

AI models can analyze complex data to predict disease progression and identify early signs of kidney damage. This allows for earlier detection and better treatment planning, making a significant impact on patient outcomes.

Terrence Sejnowski wins inaugural World Digital Technology Academy Award

Terrence Sejnowski receives Scientific Breakthrough Award for his foundational development of Boltzmann machines, providing the architectural bedrock for deep learning and generative AI. His work has had a profound impact on modern artificial intelligence and tools like ChatGPT.

AI can give as good as it gets ... or better: The moral dilemma of combative chatbots

A recent study from Lancaster University reveals that AI systems like ChatGPT can learn to mirror human impoliteness, potentially escalating into verbal violence. The research tested ChatGPT's ability to respond to real-life impolite interactions, finding it often produces more impolite behavior than humans.

Ebenbuild publishes validation study in Nature Communications Medicine demonstrating strong predictive performance of lung digital twins

A study by Ebenbuild confirms strong predictive performance of patient-specific lung digital twins, validated against clinical imaging data. The model's accuracy was demonstrated through a correlation coefficient of 0.95 with experimental data, enabling locally resolved predictions of aerosol deposition at the individual patient level.

Waves hit different on other planets

Scientists at MIT have developed a new wave model called PlanetWaves that predicts how waves will behave on planetary bodies with different liquids, atmospheres, and gravity. The model reveals that gentle winds can create massive waves on Titan, while hurricane-force winds barely move the surface of lakes on exoplanet 55-Cancri e.

Researchers simulate novel brain architecture for fast and flexible decision-making

Researchers propose a novel brain architecture for efficient processing, integrating parallel cortical and subcortical pathways. This approach may improve decision-making tasks, suggesting current AI models are missing key brain function principles.

Computational “time machine” shows solar and wind on track for 2°C target but not for 1.5°C

A new computational model developed by researchers at Chalmers University of Technology shows that onshore wind is likely to supply around 25% of global electricity by 2050 and solar reaching about 20%, aligning with the 2°C target. However, this falls short of what is required for a 1.5°C goal.

Envisioning just futures

A new IIASA-led study offers a practical approach to evaluate the distributional implications of climate action, enabling policymakers to assess justice impacts of emissions pathways. The study found that many existing mitigation scenarios are consistent with justice principles, while few explore limits on energy or meat consumption.

Integrative experiment design reveals hidden patterns in decades-old social science research

MIT Sloan researchers demonstrate a new approach to designing social science experiments by applying integrative experiment design to a longstanding question about punishment in public goods games. The study reveals how punishment's effect on collective welfare can swing from substantially harmful to substantially helpful, depending on...

New AI technology to speed drug development

Scientists at the University of Virginia Health System have developed a suite of AI-powered tools, called YuelDesign, YuelPocket and YuelBond, to transform how new drugs are created. These tools can design drug molecules tailored to fit their protein targets exactly, even accounting for protein flexibility.

Heat from traffic is contributing to rise in city temperatures, new study finds

A new study by the University of Manchester has developed a way to measure the impact of urban traffic on city temperatures, revealing that everyday vehicle use can raise temperatures in cities. The researchers found that traffic heat increases simulated air temperatures by around 0.16°C during summer and 0.35°C in winter.

UH engineer exposes structural weakness driving lithium-ion battery failure

Researchers have discovered that lithium dendrites in batteries are unexpectedly strong and brittle, causing short circuits and safety risks. The findings suggest that future battery design must change to improve safety and reliability of high-energy storage systems.

San Diego Zoo Wildlife Alliance and Scripps Institution of Oceanography partner for innovative conservation efforts

The partnership aims to boost conservation impact by harnessing advanced digital technology and biobanking techniques, focusing on three core areas: knowledge transfer, innovation, and student engagement. This collaboration seeks to address pressing environmental challenges such as biodiversity threats and emerging pathogens.