Research using machine learning to solve scientific problems. Covers AI quality control, sea-surface temperature estimation, label-free cell analysis, and cyber defense.
Researchers developed a new system called Murakkab to optimize agentic workflows for AI applications. It enables developers to describe their intent in high-level terms, automating the selection of models and tools to use, and optimizing hardware configurations in real-time.
Researchers developed a compact, cost-effective diagnostic platform combining lensfree holography and deep learning for automated HER2 scoring. The system reached 84.9% accuracy for four-level HER2 classification and 94.8% accuracy for binary scoring, effectively lowering diagnostic risks.
Researchers Anna Galler and Bettina Könighofer at Graz University of Technology will use FWF's Astra awards to identify quantum materials for future electronics and develop trustworthy AI systems. Their projects focus on protecting AI systems from risky actions and exploring new materials with unique electronic properties.
Researchers at Tohoku University have created a clearer map for searching for hydrogen storage materials, identifying key physical factors that control their performance. The study suggests adjusting geometry and lattice flexibility to raise capacity while tuning stiffness to keep equilibrium pressure near everyday conditions.
A new study from Kaiser Permanente found that combining AI mammographic risk scores with polygenic and clinical risk scores more accurately identifies women at high risk of developing breast cancer than clinical risk scores used alone. The combined model was found to improve prediction accuracy, particularly among women at highest risk.
The UN has launched an initiative calling on AI companies to publicly disclose their environmental impacts, including carbon, water and land footprint. The move comes after a report highlighted the massive electricity demand and environmental impacts of AI systems.
A University of Houston engineering professor developed a mathematical model to help decision-makers decide where to spend limited dollars on infrastructure resilience. The model accounts for real-world uncertainty and identifies critical assets to invest in, providing the greatest benefit before disaster strikes.
MIT researchers developed a new system-on-a-chip called Gleanmer, which generates highly accurate 3D maps of the robot's environment using Gaussians to represent obstacles. This approach reduces memory and power consumption by up to 99%, making it suitable for lightweight augmented reality headsets.
Researchers developed RNovA algorithm to identify new PTMs in human cells, expanding capabilities of machine learning in basic biological research. The discovery aims to advance diagnostics and broaden biologists' horizons for cancer and other diseases.
A new research direction proposes building machine-learning systems on top of AI models to detect hidden information and predict behavior. This enables users to supervise the model, control its behavior without understanding the entire mechanism.
A groundbreaking technology called Time to Move (TTM) offers unprecedented control over object and character movement in AI-generated videos. TTM eliminates the need for complex infrastructure or training on millions of videos, making AI video creation more accessible.
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.
Researchers developed an AI system called SmartTrap that uses optical tweezers to capture particles, take measurements, and load new samples autonomously. This technology accelerates the analysis of life's smallest components, potentially transforming laboratories in the near future.
A new study uses AI to identify promising chemical compounds that could develop into effective antibiotics against multi-drug resistant Neisseria gonorrhoeae. The approach has the potential to address the growing crisis of antimicrobial resistance in this fast-evolving pathogen.
A new spatial memory system allows robots to rapidly form and recall detailed mental models of large-scale environments, enabling fast and accurate object recognition. This framework combines advanced map representations with rich descriptions of the environment, enabling robots to answer complex queries in plain language.
Researchers developed BRIDGE, a multilingual benchmark that assesses large language models' understanding of clinical patient-care text. The benchmark reveals significant gaps in LLM performance on real-world clinical tasks, particularly in nuanced clinical language.
A Concordia-led team developed an AI-based method for detecting toxic online content, which outperformed existing tools in accuracy and throughput. The Proximal Policy Optimization-based Cascaded Inference System (PPO-CIS) layers scanning tasks to quickly identify harmful material.
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.
Avishek Choudhury, a WVU researcher, has won the NSF CAREER award to study how healthcare providers' trust in artificial intelligence changes over time. His goal is to humanize algorithms behind AI and improve decision-making quality and patient safety.
A robotic pet rabbit named Mía has been developed to recognize users by their voice, allowing for personalized affective stimulation in elderly care. The system uses a unique 'voice signature' that adapts to each user's speech patterns, enabling the robot to respond differently to various individuals.
Researchers at Nagoya University developed an AI tool, DiSPAH, to estimate ALS disease progression speed and identify muscle decline patterns. The study found six distinct patterns of disease progression among patients, with some experiencing rapid deterioration while others declined slowly.
Researchers create PhishLumos AI system to detect phishing campaigns by analyzing infrastructure clues, achieving 8-day faster detection than experts. The system uncovered over 190,000 new links, with 92% later flagged as malicious, outperforming content-centric approaches.
The University of Oklahoma is establishing the Oklahoma Center of ImmunoEngineering with an $11.5 million NIH award. The center will integrate wet lab science and data science to accelerate disease research. Four early-career faculty members are selected as research project leaders, and the center offers training workshops, seminars an...
A team of researchers from The University of Osaka has developed a new approach for depth reconstruction from defocus, estimating distances by analyzing blur in an image. Their method combines a coded-aperture camera with diffusion-model-based AI to accurately estimate depth and produce high-quality images.
The ACM Technology Policy Council's TechBrief examines agentic AI's legal liability, security risks, and workforce impacts. Existing frameworks fall short in addressing accountability questions, highlighting the need for defined authentication and delegation standards, robust audit trails, and sector-specific guidance.
Researchers found that friendly, choice-oriented language in AI chatbots boosts patient engagement, while aggressive phrasing and blurred human-AI boundaries put them off. Patients appreciate personalized interactions, but are wary of data security concerns.
FireANTs, an open-source algorithm, combines AI optimization and geometry to quickly match complex medical images. The new method can accomplish what took weeks in minutes, detecting subtle changes that signal disease or cognitive decline, making it practical for clinical practice.
TurboLynx, developed by POSTECH researchers, analyzes complex, interconnected data up to 184 times faster than existing systems. The engine groups similar data together and processes them collectively, reducing unnecessary memory usage and enabling efficient analytical queries.
Researchers used deep learning to model energy release during r-process nucleosynthesis in hydrodynamic simulations, gaining new insights into element formation. The results suggest that r-process heating is an important effect that should be better accounted for in future modeling.
A neural network-based machine learning model accelerates diffuse optical tomography by over a million-fold, enabling real-time diagnosis. The model accurately reproduces signals even for unseen parameter combinations, with each inference taking approximately 2 milliseconds.
A new study found that AI-powered chatbots can make vaccine-hesitant parents more likely to say they will immunize their children against HPV, but no more than standard written public health materials. Additionally, the effects of the chatbots did not last longer than those of government health materials.
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.
A new ultra-lightweight AI model, Multinex, advances low-light image enhancement by leveraging classical colour vision theory and Retinex principles. The model outperforms comparable compact systems, recovering detail and clarity from previously unusable images.
Researchers developed an AI system called OCTCube-M that can accurately identify eight different retinal diseases, including age-related macular degeneration. The technology also predicts the progression of severe forms of this condition and can infer health risks beyond the eye.
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 tool developed by Concordia researchers uses artificial intelligence to plan surgical schedules, reducing wait times and minimizing disruptions. The system can adapt to emergency surgeries while keeping non-emergency cases on schedule.
Researchers have developed a wearable sensor that reads chemical signatures of human breath to decode silent speech into text. The device uses a microscopic nanoforest to capture rapid water vapor changes, achieving 98.51% accuracy rate.
Researchers developed a novel method to predict physical phenomena without estimating parameters, reducing computational challenges. The new approach uses a multiparameter eigenvalue-problem emulator to directly predict unknown observables.
Researchers identified tens of millions of small wetlands globally and found they produce a significant impact on methane emissions. Small wetlands have been difficult to detect due to their size, but high-resolution satellite imagery has revealed their substantial contribution to the world's total non-forested wetland emissions.
Researchers have developed soft, brain-inspired electronics that can sense, store, and process information while conforming to biological tissues. These devices mimic the chemical processing of the human brain, executing complex tasks like heart rhythm classification at ultra-low voltages.
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.
IMDEA Networks has secured four SNS JU projects to develop next-generation 6G networks in Europe. The projects address key challenges including sustainability, security and industrial transformation. IMDEA Networks is leading two of the projects with principal investigators Marco Fiore and Joerg Widmer.
The global data centers powering artificial intelligence are projected to consume nearly triple the combined annual electricity use of Pakistan, Bangladesh, and Nigeria by 2030. AI's environmental cost is being systematically mismeasured, with its water footprint equaling the basic annual domestic water needs of all 1.3 billion people ...
The center will develop new phage-based treatments for antibiotic-resistant bacterial infections, predicting which phage to use for which patient and designing more effective phages. The goal is to generate unprecedented data and train AI models to identify the right phage for any patient's infection.
A new study published in the Journal of Big Data highlights the journal's emergence as a leading publication in data science and artificial intelligence research. The study found that JBD has become a central hub for high-impact research worldwide, with significant contributions from top researchers.
Researchers have developed a new citation ranking system called EDM to identify and credit scientific breakthroughs. The system uses neural language models to provide stable representations of papers and can detect simultaneous discoveries, fixing a blind spot in the original disruption index.
Binghamton University researchers have developed a new way to reduce troublesome fake information in AI chatbots, with high accuracy in identifying disease terms and drug names. The protocol harnesses multiple large language models to verify answers through 'voting', increasing confidence in the results.
The foundation has awarded fellowships to Minsoo Kim, Sahana Kuthyar, and Matthew Leventhal to investigate rare cell populations in healthy tissues and pneumonia risk in immunocompromised patients. They aim to develop new computational tools to understand biomarkers and microenvironmental influences.
Researchers at IRB Barcelona used AI to design new chemical entities that selectively target specific cell types, demonstrating superior activity compared to conventional screening strategies. The methodology, called phenotypic discovery, uses observable responses in cells rather than a specific molecular target.
A new study shows AI can generate hundreds of convincing finance research papers efficiently, but also raises concerns about the potential impact on academic community and meaning of scientific discovery. The study demonstrates how AI can accelerate research paper production while highlighting areas for improvement in peer-review systems.
A deep learning model combines knowledge from different catalyst families to identify a top-performing green hydrogen catalyst. The AI correctly predicted the activity ranking of 12 tested catalysts within a previously unexplored material family.
A new framework, SUVA, enables organizations to measure and adjust AI chatbots' social preferences, improving their performance in customer complaints and other human-AI interactions. By understanding an LLM's existing tendencies, organizations can decide whether an available model already fits its values and usage scenarios.
The Association for Computing Machinery announced three technical awards for innovations in global wireless standards, machine learning, and 3D generative AI. Erdal Arikan received the Paris Kanellakis Theory and Practice Award for his discovery of channel polarization and polar codes.
Researchers at Southwest Research Institute (SwRI) and Texas Biomedical have identified nearly two dozen antiviral compounds that could potentially treat Bundibugyo Ebolavirus. The project utilizes AI and machine learning tools to quickly identify drug candidates, with Texas Biomed set to screen the compounds in the coming weeks.
Researchers developed a neural network approach that learns to clean co-movement patterns in markets before building portfolios. The method achieved lower volatility and higher Sharpe ratios compared to traditional methods.
Scientists have demonstrated that megalibraries can design materials with specific properties, accelerating the traditional trial-and-error approach to rapidly designing and testing materials. The platform generates vast datasets needed to train AI systems to discover next-generation materials.
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 machine-learning guided lifestyle coaching program based on data collected via personal devices reduced depressive symptoms by six weeks. Participants who implemented the program experienced significant reductions in depressive symptoms and the treatment effect persisted during three months after the intervention ended.
Researchers used top Generative AI models to grade hundreds of undergraduate essays, finding that AI only matched human-awarded degree classification around half the time. The AI systems were overly sensitive to linguistic features, giving out higher marks based on essay length and vocabulary range rather than academic quality.
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.