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...
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.
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.
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.
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.
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.
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.
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%.
A new AI model called COMPASS improves the prediction of which patients are most likely to respond to cancer immunotherapy drugs, outperforming existing approaches by 8.5 percent. The model makes predictions based on tumor gene activity and provides a rationale for its output.
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.
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.
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.
A team of MIT researchers has developed a machine-learning approach that captures the diversity of atomic environments in chemically disordered materials. This allows for more accurate predictions of material properties and opens up possibilities for creating new sustainable steels and materials for aerospace, energy, and computing.
MD4SB integrates molecular dynamics tools into three European Research Infrastructures, enabling researchers to share and reuse data, and accelerate drug discovery. The project also involves pharmaceutical companies like Almirall and Sanofi.
Researchers propose a Digital Twin Optical Computing System that reduces dependence on physical hardware for task development. The DT-OCS framework enables offline simulation, training, and optimization of computational tasks, improving research efficiency and application flexibility.
Researchers have developed a generative AI model called Void-X that can predict protein-protein interactions with high accuracy, enabling the design of new biomolecules for drug discovery and synthetic biology. The model achieves predictive accuracies of 78.3% for intra-chain clusters and 68.2% for inter-chain clusters.
A study found that south-facing green walls can improve indoor thermal conditions by up to 1.7°C and enhance outdoor thermal comfort through albedo effects. Low albedo exterior surfaces also improved outdoor thermal comfort by approximately 1.5°C, while high albedo surfaces reduced indoor temperatures.
A new underwater mapping technique, Sonar-MASt3R, combines sonar and visual data to generate detailed 3D maps of environments in real-time. The system enables vehicles to navigate through cloudy water by quickly mapping the general shape of their surroundings using sonar.
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%.
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 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.
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.
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.
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.
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.
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.
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...
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.
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...
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.
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.
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.
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.
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.
Scientists are calling for a fundamental rethink of climate and biodiversity models to address the scale and complexity of crises ahead. The current models focus on incremental change rather than transformative transformations needed for a safe and just future.
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.
SourcePLOS·JournalPLOS Computational Biology·TypeExperimental study·DateMay 14, 2026
Researchers developed a new method to carry out long-term climate simulations at a fraction of the cost and time needed compared to traditional climate models. The team used a powerful climate model to train a much faster statistical emulator that can reproduce climate changes in just minutes.
A team of researchers has developed a way to precisely move tens of thousands of individual atoms within a material in minutes at room temperature. This approach uses algorithms to carefully position an electron beam and scan the beam to drive atomic motions, enabling the creation of defects with tunable functions.
The UC San Diego team developed an open-source digital twin platform called Tiny-Twin, which enables fast and realistic testing of applications over wireless networks. The platform allows researchers to test complete applications, such as video streaming, without needing specialized hardware or proprietary data.
Researchers developed AI models using electronic health records and electrocardiograms to identify individuals at elevated risk for sudden cardiac arrest. The combined EHR-EKG model correctly predicted 153 of 228 high-risk patients who experienced a cardiac arrest, bringing theoretical risk into focus.
The partnership aims to bridge the gap between fundamental research and real-world deployment of neuromorphic computing technologies. Key findings include the development of proof-of-concept applications, scalable solutions, and training programs for skills development in this rapidly emerging field.
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.
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.
A new study by TU Graz finds that women have a 60 per cent higher injury risk in car accidents compared to men. Women suffer noticeably more severe injuries even at lower collision speeds, with the risk of serious injury or death being twice as high. The study highlights the need for more realistic tests and intelligent safety systems.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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 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.
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.
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.
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.