Researchers developed a new technique combining physical laws with artificial intelligence to identify material properties even under data-scarce conditions. The approach enables rapid exploration of new materials and improves experimental efficiency while preserving reliability.
A research team at KAIST has discovered that the huntingtin protein organizes cytoskeletal microfilaments into bundles, playing a crucial role in neural connectivity. This finding is expected to contribute to understanding Huntington's disease and other neurodegenerative disorders.
A KAIST research team has developed a Frequency Switching Neuristor that autonomously adjusts its response characteristics, like the brain's intrinsic plasticity. The device achieves 27.7% less energy consumption than conventional neural networks and can compensate for partial circuit failures to resume normal operation.
Research at KAIST found that nuclear hypertrophy in cancer cells can actually suppress metastasis, rather than being a sign of worsening disease. The study identified DNA replication stress as the cause of nuclear enlargement, which can impose constraints on cancer cell potential for spread.
Researchers developed a technology combining AI analysis with optogenetics to diagnose Parkinson's disease in mice. The AI-predicted Parkinson's disease score exhibited significant differences from controls, proving sensitive in assessing severity.
A KAIST-led team has identified a gene, NR3C1, that regulates the developmental switch in astrocytes, shaping immune system reactions in adulthood. Early-life epigenetic 'memory' can predispose adult brains to excessive inflammation and disease vulnerability.
A KAIST team discovered how gallium doping improves platinum-nickel catalysts' performance and durability, retaining octahedral shape and highly active facets. This breakthrough could lead to more durable fuel cells for hydrogen-powered vehicles and clean energy technologies.
A KAIST research team has developed a wearable sensor that can analyze multiple metabolites in sweat, enabling precise monitoring of internal metabolic changes over time. The patch uses nanophotonics and microfluidics technologies to collect and analyze sweat components without labeling.
Researchers at KAIST have developed a new liquid electrolyte that significantly improves lithium-metal battery performance, enabling fast charging times of just 12 minutes and an 800km driving range. The breakthrough overcomes the long-standing dendrite problem, a major barrier to widespread adoption of lithium-metal batteries.
Researchers at KAIST have precisely clarified the operating principle of an oxide-based memory device using a multi-modal scanning probe microscope. The study reveals that oxygen defects determine the on/off state of the memory and confirms that electronic behavior also plays a role in its resistance changes.
A KAIST research team has developed a universal technology that identifies gene control targets in altered cellular gene networks and restores them. By applying an algebraic approach, they can quickly and accurately calculate how the overall cellular response would change if a specific gene were controlled.
Researchers at KAIST have developed a new time-series domain adaptation technology that allows existing AI models to detect defects accurately even when manufacturing processes or equipment change. The technology achieved up to 9.42% improvement in accuracy compared to existing methods.
Researchers identify PELOTA protein as central regulator of aging, extending lifespan through ribosome-associated quality control. Deficiency in PELOTA accelerates aging by disrupting mTOR and autophagy pathways.
A KAIST research team has developed a platform that generates specific signaling molecules in situ from a single precursor under an applied electrical signal, enabling switch-like, precise spatiotemporal control of cellular responses. The platform can produce either nitric oxide or ammonia on demand using only an electrical signal.
Researchers at KAIST have developed an AI model named BInD, which can design and optimize drug candidate molecules tailored to a protein's structure alone. The model predicts the binding mechanism between the drug and the target protein, enabling comprehensive design in one step.
Researchers have developed a next-generation wireless ophthalmic diagnostic technology incorporating an ultrathin OLED into a contact lens. This breakthrough simplifies the conventional ophthalmic diagnostic environment, allowing patients to undergo retinal function tests while wearing the lens.
Koo's team received the award for analyzing gait-related knee joint motion in healthy individuals and patients after ACL reconstruction with ALL augmentation. The study found excessive anterior translation and internal rotation, suggesting incomplete restoration of normal joint kinematics post-surgery.
The KAIST research team, led by Professor Insu Yun, won the DARPA AI Cyber Challenge with their AI-based autonomous cyber defense technology. The team scored a dominant victory, detecting and patching over 70 intentionally injected vulnerabilities in real-time.
Researchers at KAIST have developed a 'field-programmable robotic folding sheet' that can be programmed in real time according to its surroundings. The technology enables real-time shape programming and opens new possibilities in robotics.
Researchers found that adding ethane to methanotrophs' core metabolism boosts polyhydroxybutyrate (PHB) synthesis and reduces methane consumption. The study sheds new light on methanotrophic metabolism in mixed-gas environments, offering possibilities for sustainable biopolymer production.
A KAIST research team found that placental inflammation during pregnancy affects the fetus's immune system, leading to stronger allergic reactions after birth. This study presents a new possibility for early prediction and prevention of allergic diseases like pediatric asthma.
Researchers developed AI technology 'MARIOH' that can accurately reconstruct higher-order interactions from low-order interaction data, achieving up to 74% greater accuracy compared to existing methods. The model has potential applications in social network analysis, life sciences, and neuroscience.
KAIST researchers have developed a method to chemically synthesize herpotrichone, a natural substance with excellent anti-neuroinflammatory effects. The team successfully synthesized the substances using a biosynthetically inspired strategy, including the Diels-Alder reaction and hydrogen bonding.
A new sleep algorithm developed by KAIST researchers helps users build healthy sleep habits by recommending the best bedtime based on their body's circadian rhythm. The algorithm is available on Samsung Galaxy smartwatches and aims to improve sleep quality and overall well-being.
Researchers at KAIST discovered critical security vulnerabilities in smartphone communication modems, which can be exploited with a single manipulated wireless packet. The study highlights the need for standardized mobile communication modem security testing.
A KAIST research team has developed an optogenetic platform, RELISR, that enables precise spatiotemporal control over the storage and release of proteins and mRNAs. The system employs specific protein-protein and protein-RNA interactions to isolate target molecules within engineered membrane-less condensates.
KAIST researchers developed a next-generation wearable platform that utilizes ambient light as an energy source, reducing battery load by up to 86.22%. The platform integrates three complementary technologies to enable 24-hour continuous operation, including photometric, photovoltaic, and photoluminescent methods.
A Korean research team has developed a new single-atom catalyst that selectively performs only peroxidase-like reactions while maintaining high reaction efficiency. The platform enables test results to be read within minutes in near-physiological conditions, improving medical accessibility and ensuring timely treatment.
A research team at KAIST has developed a new catalyst for reducing air pollution by utilizing platinum diselenide, a two-dimensional material. The catalyst exhibits superior carbon monoxide oxidation performance across the entire temperature range compared to conventional platinum thin films.
A team of researchers at KAIST has identified a specialized gut-brain pathway for glucose sensing, which plays a central role in maintaining energy balance and metabolic homeostasis. This discovery opens up new therapeutic possibilities for metabolic diseases.
A research team elucidated the mechanism inducing plant autoimmune responses and proposed a novel strategy for cultivar improvement. The study identified defects in protein structure as the cause of hybrid necrosis, which occurs due to an abnormal reaction of immune receptors during cross-breeding between plant hybrids.
A new technology enables high-performance and efficient inference for planning based on diffusion models, improving inference-time scalability. The research achieved a 100% success rate on the giant maze-solving task, demonstrating its performance in real-time decision-making applications like intelligent robotics.
A KAIST research team has identified a new molecular mechanism in which alcohol-damaged liver cells increase reactive oxygen species, leading to cell death and inflammatory responses. They discovered that Kupffer cells act as a 'dual-function regulator' that can either promote or suppress inflammation through interactions with liver ce...
Researchers at KAIST developed a new artificial sensory nervous system that enables robots to efficiently respond to external stimuli like humans. The system mimics the functions of a living organism's sensory nervous system, allowing robots to selectively react to important or dangerous signals while ignoring safe or familiar ones.
A KAIST research team created an integrated platform that replicates brain-like layered neuronal structures using 3D printing and precisely measures neuronal activity. The platform, developed by Professors Je-Kyun Park and Yoonkey Nam, enables the simultaneous analysis of structure and function with high precision.
A KAIST team engineered a microbial strain capable of producing lutein at industrially relevant levels. They developed an electron channeling system to overcome previous limitations, achieving record-breaking 1.78 g/L of lutein production in fed-batch fermentation.
Researchers at KAIST developed a computational framework to predict key metabolic genes that can re-sensitize resistant cancer cells to treatment. This approach holds promise not only for various cancer therapies but also for treating metabolic diseases like diabetes.
Researchers from KAIST developed a high-performance, low-power NPU technology that can improve the inference performance of generative AI models like ChatGPT by over 60% while consuming approximately 44% less power compared to the latest GPUs.
Se Jin Park's SpeechSSM model generates long-duration speech without time constraints, maintaining semantic and speaker consistency. It effectively processes unbounded speech sequences by dividing data into short units and using a Non-Autoregressive audio synthesis model to generate high-quality speech.
A Korean research team used machine learning to identify a new material that can effectively remove iodate from contaminated water. The multi-metal LDH developed in this study showed exceptional adsorption performance, removing over 90% of iodate.
A KAIST research team discovered a method to enhance glioblastoma immunotherapy by restoring microbial diversity in the gut. Tryptophan supplementation activated beneficial strains of bacteria, increasing CD8 T cell infiltration into tumor tissues and improving survival rates.
A new e-textile platform developed by KAIST's research team combines 3D printing technology with advanced materials engineering to create customized training models for individual combatants. The platform uses flexible and highly durable sensors and electrodes printed directly onto textile substrates, enabling precise movement and huma...
A new recycling technology has been developed to turn used tires into raw materials for rubber and nylon, achieving high selectivity of up to 92% and a yield of 82%. The process uses dual catalysis to convert waste rubber into valuable chemicals.
Researchers at KAIST have created a new Li-Fi platform that offers speeds up to 224 Gbps and enhances security through on-device encryption. The device uses eco-friendly quantum dots and demonstrates improved brightness and efficiency, opening up new possibilities for ultra-high-speed data communication.
Researchers at KAIST have developed a technology to enhance creative generation of AI generative models like Stable Diffusion, generating novel and useful images. The algorithm amplifies internal feature maps to boost creativity without new training, outperforming existing methods in novelty and utility.
Researchers at KAIST develop a 'pedestrian-friendly smart window' technology that reduces heating and cooling energy consumption in urban buildings while resolving light pollution issues. The RECM system operates in three modes, allowing for real-time adjustment of light and heat transmission.
Researchers developed a technology that precisely analyzes 21 types of reactants simultaneously using high-resolution fluorine nuclear magnetic resonance spectroscopy. This breakthrough contributes to new drug development and catalyst optimization in AI-driven autonomous synthesis.
Researchers at KAIST have developed a groundbreaking technology capable of selectively acetylating specific RNA molecules within the human body using the CRISPR-Cas13 system. This breakthrough enables precise, programmable control of RNA function and is expected to open new avenues in RNA-based therapeutic development.
The virtual teaching assistant (VTA) provides personalized feedback to individual students even in large-scale classes. The system, which automatically vectorizes a large volume of course materials and uses them as the basis for answering students' questions, has been shown to significantly reduce the burden on TAs.
Researchers at KAIST developed a quadrupedal navigation system that enables the robot to reach its target destination quickly and safely in complex terrain. Inspired by cat's paw placement, they significantly reduced computational complexity.