%41%43%54
-
KAIST Researchers Unveil Technical Principles Behind Antibacterial Graphene Toothbrushes with 10 Million Units Sold
< (From left) Professor Hyun Jung Chung , Ph.D candidate Ju Yeon Chung, Ph.D candidate Sujin Cha, Professor Sang Ouk Kim >
Hygiene in everyday items that touch the body—such as clothing, masks, and toothbrushes—is critically important. The underlying principle of how graphene selectively eliminates only bacteria has now been revealed. A KAIST research team has presented the potential for a next-generation antibacterial material that is safe for the human body and capable of replacing antibiotics.
KAIST announced on March 25th that a joint research team, led by Professor Sang Ouk Kim from the Department of Materials Science and Engineering and Professor Hyun Jung Chung from the Department of Biological Sciences, has identified the mechanism by which Graphene Oxide (GO) exhibits powerful antibacterial effects against bacteria while remaining harmless to human cells. Graphene oxide is a nanomaterial consisting of an atomic level carbon layer (graphene) with oxygen attached; it is characterized by its ability to mix well with water and implement various functions.
This study is highly significant as it provides molecular-level proof of graphene's antibacterial action, which had not been clearly understood until now.
The research team confirmed that graphene oxide performs "selective antibacterial action" by attaching to and destroying only the membranes of bacteria, much like a magnet attaches only to specific metals, while leaving human cells untouched. This occurs because the oxygen functional groups on the surface of graphene oxide selectively bind with a specific component (POPG) found only in bacterial cell membranes. Simply put, it recognizes a "target" present only in bacterial membranes to attach and destroy the structure. In this context, phospholipids are fatty components that make up the membrane surrounding a cell, and POPG is a component primarily present in bacteria.
< Schematic diagram of the selective interaction between graphene oxide and cell membranes >
< Identification of selective interaction mechanisms at the molecular level through microscopic and chemical analysis of artificial lipid vesicles mimicking cell membranes >
Nanofibers applying this principle effectively inhibited the growth of various pathogenic bacteria, including superbugs resistant to antibiotics. Animal experiments also confirmed its effectiveness in promoting wound healing without inducing inflammation.
< Verification of antibacterial and wound healing enhancement effects in a porcine infected wound model >
Furthermore, fibers using this material maintained their antibacterial functions even after multiple washes, showing potential for use in various industrial fields such as apparel and medical textiles.
This technology is already being applied to consumer products. The graphene antibacterial toothbrush, released through the original patents of the faculty-led startup 'Materials Creation Co., Ltd.,' has sold over 10 million units, proving its commercial viability. Additionally, GrapheneTex—textile materiala incorporating this technology—was used in the uniforms of the Taekwondo demonstration team at the 2024 Paris Olympics and is expected to play an active role in functional sportswear at upcoming international sporting events like the 2026 Asian Games.
< Commercially available graphene toothbrush >
< Graphene material image (AI-generated image) >
Professor Sang Ouk Kim explained, "This study is an example of scientifically uncovering why graphene can selectively kill bacteria while remaining safe for the human body." He emphasized, "By utilizing this principle, we can expand beyond safe clothing without harsh chemicals to an infinite range of applications, including wearable devices and medical textile systems."
Sujin Cha (PhD program, Department of Materials Science and Engineering) and Ju Yeon Chung (Integrated MS/PhD program, Department of Biological Sciences) participated as first authors. Professor Hyun Jung Chung participated as a co-corresponding author. The research was published on March 2nd in the prestigious materials science journal, Advanced Functional Materials.
※ Paper Title: Biocompatible but Antibacterial Mechanism of Graphene Oxide for Sustainable Antibiotics, DOI: 10.1002/adfm.202313583
Additionally, Nanowerk (http://www.nanowerk.com/), a global portal for nanotechnology, featured these findings as a 'Spotlight' titled "Graphene oxide destroys bacteria without harming human tissue."
This research was conducted with support from the 'Nano/Material Technology Development (R&D)' program, the 'Individual Basic Research' program, and the 'Mid-Career Researcher Support Program' funded by the Ministry of Science and ICT.
2026.03.26 View 3953 -
World’s First AI-Managed Unmanned Factory Implemented... Construction of Physical AI KAIROS
< Integrated Operation of Heterogeneous Logistics Robot Systems >
KAIST announced on March 23rd that Professor Young Jae Jang's team from the Department of Industrial and Systems Engineering has constructed ‘KAIROS’ (KAIST AI Robot Orchestration Systems), a physical AI testbed that integrates and controls heterogeneous robots, sensors, facilities, and digital twins into a single system.
KAIROS is a 100% unmanned factory platform based on physical AI and is the first integrated testbed of its kind in Korea, developed with support from the Ministry of Science and ICT (MSIT). It is particularly noteworthy as a domestic integrated solution aimed at exporting "Dark Factories" in the future.
The most significant feature of KAIROS is its structure, which integrates and controls various factory equipment through a single AI agent-based Operating System (OS). While existing factory automation was operated around individual devices, KAIROS integrates Autonomous Mobile Robots (AMR), humanoid robots, collaborative robots, and automation facilities into a single intelligent platform. Through this, the concept of ‘Physical AI-based factory operation’—where the entire factory is operated like a single AI system—has been realized.
The core of this testbed is the 100% domestic integration of the entire process from sensors and control to data processing. By integrating key elements of a Dark Factory—including logistics robots (AMR), OHT, 3D shuttles, humanoid robots, collaborative robots, industrial sensors and PC controllers, wireless charging systems, digital twins and simulations, and AI-based integrated control and safety management systems—using domestic technology, the project has replaced factory automation equipment and software that were heavily dependent on foreign technology and laid the foundation for a ‘K-Manufacturing Factory Export Model.’
As part of the Physical AI Pre-verification Project, the MSIT has supported the establishment of a demonstration lab within the KAIST Industrial Management Building. On March 23, Vice Minister Bae Gyeong-hoon (Minister of Science and ICT) visited KAIST to announce the National Physical AI Strategy (Draft) and unveil the KAIROS-based Dark Factory demonstration site.
At the event, the factory operating system of the KAIST demonstration lab, joint physical AI demonstration results with Chonbuk National University, and the direction of the ‘Team Korea Physical AI (TK-PAI)’ alliance—a cooperative structure of domestic companies—were discussed.
< KAIROS Operation Plan Announcement >
< KAIROS Demonstration >
< KAIROS Factory Site >
KAIST plans to further advance the next-generation factory operating system (OS), covering the design, construction, and operation of Dark Factories through KAIROS, and to develop simulation and virtual verification environments.
In addition, the university intends to utilize the platform as a testing and evaluation site where domestic robot and automation companies can pre-verify highly reliable equipment, thereby increasing industrial applicability. Furthermore, the goal is to develop physical AI-based Dark Factory solutions capable of competing with global companies such as Siemens (Germany), FANUC (Japan), and Yaskawa (Japan) to pursue entry into the global market.
Kwang Hyung Lee, President of KAIST, stated, “KAIROS is the beginning of a new industrial paradigm where AI directly operates factories. KAIST will lead manufacturing innovation based on physical AI and contribute to ensuring South Korea’s leadership in global industrial competition.”
Professor Young Jae Jang, who led the construction of KAIROS, explained, “KAIROS goes beyond individual automation technologies to implement the concept of a factory operating system (OS) that integrates diverse robots and facilities into one system. It will serve as a foundation for domestic companies to verify physical AI technologies applicable to actual industrial sites and expand into the global market.”
2026.03.23 View 2566 -
KAIST, AI judges manufacturing beyond craftsmanship and language barriers
<(From Left) M.S candidate Inhyo Lee, Ph.D candidate Heekyu Kim, Ph.D candidate joonyoung Kim, Professor Seunghwa Ryu>
Most of the plastic products we use are made through injection molding, a process in which molten plastic is injected into a mold to mass-produce identical items. However, even slight changes in conditions can lead to defects, so the process has long relied on the intuition of highly skilled workers. Now, KAIST researchers have proposed an AI-based solution that autonomously optimizes processes and transfers manufacturing knowledge, addressing concerns that expertise could be lost due to the retirement of skilled workers and the increase in foreign labor.
KAIST (President Kwang Hyung Lee) announced on the 22nd of December that a research team led by Professor Seunghwa Ryu from the Department of Mechanical Engineering · InnoCORE PRISM-AI Center has, for the first time in the world, developed generative AI technology that autonomously optimizes injection molding processes, along with an LLM-based knowledge transfer system that makes on-site expertise accessible to anyone. The team also reported that these achievements were published consecutively in an internationally renowned journal.
The first achievement is a generative AI–based process inference technology that automatically infers optimal process conditions based on environmental changes or quality requirements. Previously, whenever temperature, humidity, or desired quality levels changed, skilled workers had to rely on trial and error to readjust conditions.
The research team implemented a diffusion model–based approach that reverse-engineers process conditions satisfying target quality requirements, using environmental data and process parameters collected over several months from an actual injection molding factory.
In addition, the team built a surrogate model that substitutes for actual production, enabling quality prediction without running the real process. As a result, they achieved an error rate of just 1.63%, significantly lower than the 23~44% error rates of representative existing technologies such as GAN* and VAE** models traditionally used for process prediction. Experiments applying the AI-generated conditions to real processes confirmed successful production of acceptable products, demonstrating practical applicability.
*GAN (Generative Adversarial Network): a method in which two AI models compete with each other to generate data
**VAE (Variational Autoencoder): a method that compresses and learns common patterns in data and then reconstructs them
<Figure 1. Generative AI–Based Process Reasoning Technology>
The second achievement is the IM-Chat, an LLM-based knowledge transfer system designed to address skilled worker retirement and multilingual work environments. IM-Chat is a multi-agent AI system that combines large language models (LLMs) with retrieval-augmented generation (RAG), serving as an AI assistant for manufacturing sites by providing appropriate solutions to problems encountered by novice or foreign workers.
When a worker asks a question in natural language, the AI understands it and, if necessary, automatically calls the generative process inference AI, simultaneously providing optimal process condition calculations along with relevant standards and background explanations.
For example, when asked, “What is the appropriate injection pressure when the factory humidity is 43.5%?”, the AI calculates the optimal condition and presents the supporting manual references as well. With support for multilingual interfaces, foreign workers can receive the same level of decision-making support.
This research is regarded as a core manufacturing AI transformation (AX) technology that can be extended beyond injection molding to molds, presses, extrusion, 3D printing, batteries, bio-manufacturing, and other industries.
In particular, the work is significant in that it presents a paradigm for autonomous manufacturing AI, integrating generative AI and LLM agents through a Tool-Calling approach*, enabling AI to make its own judgments and invoke necessary functions.
*Tool-Calling approach: a method in which AI autonomously calls and uses the functions or programs required for a given situation
<Figure 2. Large Language Model–Based Multilingual Knowledge Transfer Multi-Agent IM-Chat>
<Figure 3. Example of Operation of the Large Language Model (LLM)–Based Multilingual Knowledge Transfer Multi-Agent IM-Chat>
<Figure 4. Illustration of the Application of an LLM-Based Multilingual Knowledge Transfer Multi-Agent IM-Chat (AI-Generated)>
Professor Seunghwa Ryu explained, “This is a case where we addressed fundamental problems in manufacturing in a data-driven way by combining AI that autonomously optimizes processes with LLMs that make on-site knowledge accessible to anyone,” adding, “We will continue expanding this approach to various manufacturing processes to accelerate intelligence and autonomy across the industry.”
This research involved doctoral candidates Junhyeong Lee, Joon-Young Kim, and Heekyu Kim from the Department of Mechanical Engineering as co–first authors, with Professor Seunghwa Ryu as the corresponding author. The results were published consecutively in the April and December issues of Journal of Manufacturing Systems (JCR 1/69, IF 14.2), the world’s top-ranked international journal in engineering and industrial fields.
※ Paper 1: “Development of an Injection Molding Production Condition Inference System Based on Diffusion Model,” DOI: https://doi.org/10.1016/j.jmsy.2025.01.008 ※ Paper 2: “IM-Chat: A multi-agent LLM framework integrating tool-calling and diffusion modeling for knowledge transfer in injection molding industry,” DOI: https://doi.org/10.1016/j.jmsy.2025.11.007
This research was supported by the Ministry of Science and ICT, the Ministry of SMEs and Startups, and the Ministry of Trade, Industry and Energy.
2025.12.22 View 2343 -
A KAIST team develops the world's first modular co-culture platform for the one-pot production of rainbow-colored bacterial cellulose.
<(From Left) Distinguished Professor Sang Yup Lee, Ph.D candidate Pingxin Lin, Ph.D candiate Zhou Hengrui>
The integration of systems metabolic engineering with co-culture strategies that couples bacterial cellulose production with natural colorant biosynthesis enabled the one-pot generation of rainbow-colored bacterial cellulose, establishing a sustainable biomanufacturing platform that can replace petroleum-based textiles and eliminate chemical dyeing processes.
A research group at KAIST has successfully developed a modular co-culture platform for the one-pot production of rainbow-colored bacterial cellulose. The team, led by Distinguished Professor Sang Yup Lee from the Department of Chemical and Biomolecular Engineering, engineered Komagataeibacter xylinus for bacterial cellulose synthesis and Escherichia coli for natural colorants overproduction. A co-culture of these engineered strains enabled the in situ coloration of bacterial cellulose. This research offers a versatile platform for producing living materials in multiple colors, and provides new opportunities for sustainable textiles, wearable biomaterials, and functional living materials that combine optical and structural properties beyond the reach of conventional textile technologies.
Bacterial cellulose is an attractive and biodegradable alternative to petroleum-derived fabrics due to its high purity, mechanical strength, and water-retention properties. However, the limited color range of bacterial cellulose, which is typically white, has limited its broader application in the textile industry, where more vibrant colored fabrics are increasingly desired. Conventional dyeing methods rely on petroleum-based colorants and toxic reagents, creating environmental and processing challenges. These challenges have driven the demand for alternative production methods.
To address these issues, KAIST researchers, including Ph.D. Candidate Hengrui Zhou, Ph.D. Candidate Pingxin Lin, Professor Ki Jun Jeong, and Distinguished Professor Sang Yup Lee, combined systems metabolic engineering with co-culture strategies to develop a bio-based route that integrates bacterial cellulose formation with natural pigment synthesis, enabling the production of colored living materials in a single step without additional chemical processing.
The team’s work, entitled “One-pot production of colored bacterial cellulose,” was published in Trends in Biotechnology on November 12,2025.
This research details the one-pot production of multicolored bacterial cellulose using a modular co-culture platform that integrates a bacterial cellulose-overproducing K. xylinus strain with natural colorant-producing E. coli strains. The team focused on addressing the limitations in bacterial cellulose coloration caused by environmental challenges and complex processing requirements. By employing vesicle engineering and optimizing co-culture parameters, the researchers achieved one-pot production of red, orange, yellow, green, blue, navy, and purple bacterial cellulose, eliminating the need for external dyes and toxic chemical treatments.
To enhance dyeing efficiency, E. coli strains were engineered for the overproduction and secretion of natural colorants. It was determined that the intracellular accumulation of these pigments disrupts cellular metabolism and physiology, thereby inhibiting their production. To overcome this limitation, vesicle engineering has emerged as a key strategy to mitigate these cytotoxic effects, including the induction of inner- and outer-membrane vesicles and the modulation of cell morphology, enabling the more efficient secretion of colorants and increased overall production. The engineered E. coli strains were optimized in fed-batch fermentation, achieving record-breaking production of 16.92 ± 0.10 g/L of deoxyviolacein, 8.09 ± 0.17 g/L of violacein, 1.82 ± 0.07 g/L of proviolacein, and 936.25 ± 9.70 mg/L of prodeoxyviolacein, the highest reported titers to date for all four violacein derivatives.
< Figure 1. Rainbow-colored bacterial cellulose (microbial fiber) with applied color >
A co-culture platform combining the K. xylinus with E. coli strains was further developed and optimized, enabling the in situ one-pot coloration of bacterial cellulose in vibrant green, blue, navy, and purple. Fed-batch fermentation further improved the performance of the platform, achieving the world-first one-pot production of multicolored bacterial cellulose on a larger scale. To expand the bacterial cellulose color palette, engineered carotenoid-producing E. coli strains were incorporated, enabling the successful synthesis of red, orange, and yellow bacterial cellulose. This milestone demonstrates the potential of microbial fermentation as a sustainable alternative to petroleum-based textile processes.
“We can anticipate that this microbial cell factory-based one-pot production of rainbow-colored bacterial cellulose has the potential to replace current petroleum-based textile processes,” said Ph.D. Candidate Hengrui Zhou. “The systems metabolic engineering strategies developed in this study could be broadly applied for the production of diverse sustainable textiles, wearable biomaterials, and functional living materials that combine optical and structural properties beyond the capabilities of conventional textile technologies.” He added, “This platform reduces the environmental impact while greatly expanding design possibilities. Beyond serving as a proof-of-concept, this technology offers a promising route toward scalable, eco-friendly fabrics with in situ coloration. Its modular design allows the incorporation of diverse natural colorant pathways, enabling the creation of living materials in multiple colors.”
< Figure 2. Schematic of a microbe-based platform for one-step production of rainbow-colored bacterial cellulose >
“As demand for sustainable textiles and living materials continues to grow, we expect that the integrated biomanufacturing platform developed here will play a pivotal role in producing diverse functional biomaterials with additional design possibilities in a single step, without additional chemical processing,” explained Distinguished Professor Sang Yup Lee.
This work was supported by the Development of Next-generation Biorefinery Platform Technologies for Leading Bio-based Chemicals Industry project (2022M3J5A1056072) and the Development of Platform Technologies of Microbial Cell Factories for the Next-generation Biorefineries project (2022M3J5A1056117) from the National Research Foundation supported by the Korean Ministry of Science and ICT.
Source:
Hengrui Zhou (1st), Pingxin Lin (2nd), Ki Jun Jeong (3rd), and Sang Yup Lee (Corresponding). “One-pot production of colored bacterial cellulose”. Trends in Biotechnology (Published) doi: 10.1016/j.tibtech.2025.09.019
2025.11.19 View 3149 -
KAIST Develops Multimodal AI That Understands Text and Images Like Humans
<(From Left) M.S candidate Soyoung Choi, Ph.D candidate Seong-Hyeon Hwang, Professor Steven Euijong Whang>
Just as human eyes tend to focus on pictures before reading accompanying text, multimodal artificial intelligence (AI)—which processes multiple types of sensory data at once—also tends to depend more heavily on certain types of data. KAIST researchers have now developed a new multimodal AI training technology that enables models to recognize both text and images evenly, enabling far more accurate predictions.
KAIST (President Kwang Hyung Lee) announced on the 14th that a research team led by Professor Steven Euijong Whang from the School of Electrical Engineering has developed a novel data augmentation method that enables multimodal AI systems—those that must process multiple data types simultaneously—to make balanced use of all input data.
Multimodal AI combines various forms of information, such as text and video, to make judgments. However, AI models often show a tendency to rely excessively on one particular type of data, resulting in degraded prediction performance.
To solve this problem, the research team deliberately trained AI models using mismatched or incongruent data pairs. By doing so, the model learned to rely on all modalities—text, images, and even audio—in a balanced way, regardless of context.
The team further improved performance stability by incorporating a training strategy that compensates for low-quality data while emphasizing more challenging examples. The method is not tied to any specific model architecture and can be easily applied to various data types, making it highly scalable and practical.
<Model Prediction Changes with a Data-Centric Multimodal AI Training Framework>
Professor Steven Euijong Whang explained, “Improving AI performance is not just about changing model architectures or algorithms—it’s much more important how we design and use the data for training.” He continued, “This research demonstrates that designing and refining the data itself can be an effective approach to help multimodal AI utilize information more evenly, without becoming biased toward a specific modality such as images or text.”
The study was co-led by doctoral student Seong-Hyeon Hwang and master’s student Soyoung Choi, with Professor Steven Euijong Whang serving as the corresponding author. The results will be presented at NeurIPS 2025 (Conference on Neural Information Processing Systems), the world’s premier conference in the field of AI, which will be held this December in San Diego, USA, and Mexico City, Mexico.
※ Paper title: “MIDAS: Misalignment-based Data Augmentation Strategy for Imbalanced Multimodal Learning,” Original paper: https://arxiv.org/pdf/2509.25831
The research was supported by the Institute for Information & Communications Technology Planning & Evaluation (IITP) under the projects “Robust, Fair, and Scalable Data-Centric Continual Learning” (RS-2022-II220157) and “AI Technology for Non-Invasive Near-Infrared-Based Diagnosis and Treatment of Brain Disorders” (RS-2024-00444862).
2025.10.14 View 4189 -
KAIST Wins Bid for ‘Physical AI Core Technology Demonstration’ Pilot Project
KAIST (President Kwang Hyung Lee) announced on the 28th of August that, together with Jeonbuk State, Jeonbuk National University, and Sungkyunkwan University, it has jointly won the Ministry of Science and ICT’s pilot project for the “Physical AI Core Technology Proof of Concept (PoC)”, with KAIST serving as the overall research lead. The consortium also plans to participate in a full-scale demonstration project that is expected to reach a total scale of 1 trillion KRW in the future.
In this project, KAIST led the research planning under the theme of “Collaborative Intelligence Physical AI.” Based on this, Jeonbuk National University and Jeonbuk State will carry out joint research and establish a collaborative intelligence physical AI industrial ecosystem within the province. The pilot project will begin on September 1 this year and will run until the end of the year over the next five years. Through this effort, Jeonbuk State aims to be built into a global hub for physical AI.
KAIST will take charge of developing original research technologies, creating a research environment through the establishment of a testbed, and promoting industrial diffusion. Professor Young Jae Jang of the Department of Industrial and Systems Engineering at KAIST, who is the overall project director, has been leading research on collaborative intelligence physical AI since 2016. His “Collaborative Intelligence-Based Smart Manufacturing Innovation Technology” was selected as one of KAIST’s “Top 10 Research Achievements” in 2019.
“Physical AI” refers to cutting-edge artificial intelligence technology that enables physical devices such as robots, autonomous vehicles, and factory automation equipment to perform tasks without human instruction by understanding spatiotemporal concepts.
In particular, collaborative intelligence physical AI is a technology in which numerous robots and automated devices in a factory environment work together to achieve goals. It is attracting attention as a key foundation for realizing “dark factories” in industries such as semiconductors, secondary batteries, and automobile manufacturing.
Unlike existing manufacturing AI, this technology does not necessarily require massive amounts of historical data. Through real-time, simulation-based learning, it can quickly adapt even to manufacturing environments with frequent changes and has been deemed a next-generation technology that overcomes the limitations of data dependency.
Currently, the global AI industry is led by LLMs that simulate linguistic intelligence. However, physical AI must go beyond linguistic intelligence to include spatial intelligence and virtual environment learning, requiring the organic integration of hardware such as robots, sensors, and motors with software. As a manufacturing powerhouse, Korea is well-positioned to build such an ecosystem and seize the opportunity to lead global competition.
In fact, in April 2025, KAIST won first place at INFORMS (Institute for Operations Research and the Management Sciences), the world’s largest industrial engineering society, with its case study on collaborative intelligence physical AI, beating MIT and Amazon. This achievement is recognized as proof of Korea’s global competitiveness in the physical AI technology realm.
Professor Young Jae Jang, KAIST’s overall project director, said, “Winning this large-scale national project is the result of KAIST’s collaborative intelligence physical AI research capabilities accumulated over the past decade being recognized both domestically and internationally. This will be a turning point for establishing Korea’s manufacturing industry as a global leading ‘Physical AI Manufacturing Innovation Model.’”
KAIST President Kwang Hyung Lee emphasized that “KAIST is taking on the role of leading not only academic research but also the practical industrialization of national strategic technologies. Building on this achievement, we will collaborate with Jeonbuk National University and Jeonbuk State to develop Korea into a world-class hub for physical AI innovation.”
Through this project, KAIST, Jeonbuk National University, and Jeonbuk State plan to develop Korea into a global industrial hub for physical AI.
2025.08.29 View 4256 -
KAIST Develops AI that Automatically Detects Defects in Smart Factory Manufacturing Processes Even When Conditions Change
Recently, defect detection systems using artificial intelligence (AI) sensor data have been installed in smart factory manufacturing sites. However, when the manufacturing process changes due to machine replacement or variations in temperature, pressure, or speed, existing AI models fail to properly understand the new situation and their performance drops sharply. KAIST researchers have developed AI technology that can accurately detect defects even in such situations without retraining, achieving performance improvements up to 9.42%. This achievement is expected to contribute to reducing AI operating costs and expanding applicability in various fields such as smart factories, healthcare devices, and smart cities.
KAIST (President Kwang Hyung Lee) announced on the 26th of August that a research team led by Professor Jae-Gil Lee from the School of Computing has developed a new “time-series domain adaptation” technology that allows existing AI models to be utilized without additional defect labeling, even when manufacturing processes or equipment change.
Time-series domain adaptation technology enables AI models that handle time-varying data (e.g., temperature changes, machine vibrations, power usage, sensor signals) to maintain stable performance without additional training, even when the training environment (domain) and the actual application environment differ.
Professor Lee’s team paid attention to the fact that the core problem of AI models becoming confused by environmental (domain) changes lies not only in differences in data distribution but also in changes in defect occurrence patterns (label distribution) themselves. For example, in semiconductor wafer processes, the ratio of ring-shaped defects and scratch defects may change due to equipment modifications.
The research team developed a method for decomposing new process sensor data into three components—trends, non-trends, and frequencies—to analyze their characteristics individually. Just as humans detect anomalies by combining pitch, vibration patterns, and periodic changes in machine sounds, AI was enabled to analyze data from multiple perspectives.
In other words, the team developed TA4LS (Time-series domain Adaptation for mitigating Label Shifts) technology, which applies a method of automatically correcting predictions by comparing the results predicted by the existing model with the clustering information of the new process data. Through this, predictions biased toward the defect occurrence patterns of the existing process can be precisely adjusted to match the new process.
In particular, this technology is highly practical because it can be easily combined like an additional plug-in module inserted into existing AI systems without requiring separate complex development. That is, regardless of the AI technology currently being used, it can be applied immediately with only simple additional procedures.
In experiments using four benchmark datasets of time-series domain adaptation (i.e., four types of sensor data in which changes had occurred), the research team achieved up to 9.42% improvement in accuracy compared to existing methods.[TT1]
Especially when process changes caused large differences in label distribution (e.g., defect occurrence patterns), the AI demonstrated remarkable performance improvement by autonomously correcting and distinguishing such differences. These results proved that the technology can be used more effectively without defects in environments that produce small batches of various products, one of the main advantages of smart factories.
Professor Jae-Gil Lee, who supervised the research, said, “This technology solves the retraining problem, which has been the biggest obstacle to the introduction of artificial intelligence in manufacturing. Once commercialized, it will greatly contribute to the spread of smart factories by reducing maintenance costs and improving defect detection rates.”
This research was carried out with Jihye Na, a Ph.D. student at KAIST, as the first author, with Youngeun Nam, a Ph.D. student, and Junhyeok Kang, a researcher at LG AI Research, as co-authors. The research results were presented in August 2025 at KDD (the ACM SIGKDD Conference on Knowledge Discovery and Data Mining), the world’s top academic conference in artificial intelligence and data.
※Paper Title: “Mitigating Source Label Dependency in Time-Series Domain Adaptation under Label Shifts”
※DOI: https://doi.org/10.1145/3711896.3737050
This technology was developed as part of the research outcome of the SW Computing Industry Original Technology Development Program’s SW StarLab project (RS-2020-II200862, DB4DL: Development of Highly Available and High-Performance Distributed In-Memory DBMS for Deep Learning), supported by the Ministry of Science and ICT and the Institute for Information & Communications Technology Planning & Evaluation (IITP).
2025.08.26 View 4342 -
KAIST Develops World’s First Wireless OLED Contact Lens for Retinal Diagnostics
<ID-style photograph against a laboratory background featuring an OLED contact lens sample (center), flanked by the principal authors (left: Professor Seunghyup Yoo ; right: Dr. Jee Hoon Sim). Above them (from top to bottom) are: Professor Se Joon Woo, Professor Sei Kwang Hahn, Dr. Su-Bon Kim, and Dr. Hyeonwook Chae>
Electroretinography (ERG) is an ophthalmic diagnostic method used to determine whether the retina is functioning normally. It is widely employed for diagnosing hereditary retinal diseases or assessing retinal function decline.
A team of Korean researchers has developed a next-generation wireless ophthalmic diagnostic technology that replaces the existing stationary, darkroom-based retinal testing method by incorporating an “ultrathin OLED” into a contact lens. This breakthrough is expected to have applications in diverse fields such as myopia treatment, ocular biosignal analysis, augmented-reality (AR) visual information delivery, and light-based neurostimulation.
On the 12th, KAIST (President Kwang Hyung Lee) announced that a research team led by Professor Seunghyup Yoo from the School of Electrical Engineering, in collaboration with Professor Se Joon Woo of Seoul National University Bundang Hospital (Director Jeong-Han Song), Professor Sei Kwang Hahn of POSTECH (President Sung-Keun Kim) and CEO of PHI Biomed Co., and the Electronics and Telecommunications Research Institute (ETRI, President Seungchan Bang) under the National Research Council of Science & Technology (NST, Chairman Youngshik Kim), has developed the world’s first wireless contact lens-based wearable retinal diagnostic platform using organic light-emitting diodes (OLEDs).
<Figure 1. Schematic and photograph of the wireless OLED contact lens>
This technology enables ERG simply by wearing the lens, eliminating the need for large specialized light sources and dramatically simplifying the conventional, complex ophthalmic diagnostic environment.
Traditionally, ERG requires the use of a stationary Ganzfeld device in a dark room, where patients must keep their eyes open and remain still during the test. This setup imposes spatial constraints and can lead to patient fatigue and compliances challenges.
To overcome these limitations, the joint research team integrated an ultrathin flexible OLED —approximately 12.5 μm thick, or 6–8 times thinner than a human hair— into a contact lens electrode for ERG. They also equipped it with a wireless power receiving antenna and a control chip, completing a system capable of independent operation.
For power transmission, the team adopted a wireless power transfer method using a 433 MHz resonant frequency suitable for stable wireless communication. This was also demonstrated in the form of a wireless controller embedded in a sleep mask, which can be linked to a smartphone —further enhancing practical usability.
<Figure 2. Schematic of the electroretinography (ERG) testing system using a wireless OLED contact lens and an example of an actual test in progress>
While most smart contact lens–type light sources developed for ocular illumination have used inorganic LEDs, these rigid devices emit light almost from a single point, which can lead to excessive heat accumulation and thus usable light intensity. In contrast, OLEDs are areal light sources and were shown to induce retinal responses even under low luminance conditions. In this study, under a relatively low luminance* of 126 nits, the OLED contact lens successfully induced stable ERG signals, producing diagnostic results equivalent to those obtained with existing commercial light sources.
*Luminance: A value indicating how brightly a surface or screen emits light; for reference, the luminance of a smartphone screen is about 300–600 nits (can exceed 1000 nits at maximum).
Animal tests confirmed that the surface temperature of a rabbit’s eye wearing the OLED contact lens remained below 27°C, avoiding corneal heat damage, and that the light-emitting performance was maintained even in humid environments—demonstrating its effectiveness and safety as an ERG diagnostic tool in real clinical settings.
Professor Seunghyup Yoo stated that “integrating the flexibility and diffusive light characteristics of ultrathin OLEDs into a contact lens is a world-first attempt,” and that “this research can help expand smart contact lens technology into on-eye optical diagnostic and phototherapeutic platforms, contributing to the advancement of digital healthcare technology.”
< Wireless operation of the OLED contact lens >
Jee Hoon Sim, Hyeonwook Chae, and Su-Bon Kim, PhD researchers at KAIST, played a key role as co-first authors alongside Dr. Sangbaie Shin of PHI Biomed Co.. Corresponding authors are Professor Seunghyup Yoo (School of Electrical Engineering, KAIST), Professor Sei Kwang Hahn (Department of Materials Science and Engineering, POSTECH), and Professor Se Joon Woo (Seoul National University Bundang Hospital). The results were published online in the internationally renowned journal ACS Nano on May 1st.
● Paper title: Wireless Organic Light-Emitting Diode Contact Lenses for On-Eye Wearable Light Sources and Their Application to Personalized Health Monitoring
● DOI: https://doi.org/10.1021/acsnano.4c18563
● Related video clip: http://bit.ly/3UGg6R8
< Close-up of the OLED contact lens sample >
2025.08.12 View 5354 -
KAIST Develops AI ‘MARIOH’ to Uncover and Reconstruct Hidden Multi-Entity Relationships
<(From Left) Professor Kijung Shin, Ph.D candidate Kyuhan Lee, and Ph.D candidate Geon Lee>
Just like when multiple people gather simultaneously in a meeting room, higher-order interactions—where many entities interact at once—occur across various fields and reflect the complexity of real-world relationships. However, due to technical limitations, in many fields, only low-order pairwise interactions between entities can be observed and collected, which results in the loss of full context and restricts practical use. KAIST researchers have developed the AI model “MARIOH,” which can accurately reconstruct* higher-order interactions from such low-order information, opening up innovative analytical possibilities in fields like social network analysis, neuroscience, and life sciences.
*Reconstruction: Estimating/reconstructing the original structure that has disappeared or was not observed.
KAIST (President Kwang Hyung Lee) announced on the 5th that Professor Kijung Shin’s research team at the Kim Jaechul Graduate School of AI has developed an AI technology called “MARIOH” (Multiplicity-Aware Hypergraph Reconstruction), which can reconstruct higher-order interaction structures with high accuracy using only low-order interaction data.
Reconstructing higher-order interactions is challenging because a vast number of higher-order interactions can arise from the same low-order structure.
The key idea behind MARIOH, developed by the research team, is to utilize multiplicity information of low-order interactions to drastically reduce the number of candidate higher-order interactions that could stem from a given structure.
In addition, by employing efficient search techniques, MARIOH quickly identifies promising interaction candidates and uses multiplicity-based deep learning to accurately predict the likelihood that each candidate represents an actual higher-order interaction.
<Figure 1. An example of recovering high-dimensional relationships (right) from low-dimensional paper co-authorship relationships (left) with 100% accuracy, using MARIOH technology.>
Through experiments on ten diverse real-world datasets, the research team showed that MARIOH reconstructed higher-order interactions with up to 74% greater accuracy compared to existing methods.
For instance, in a dataset on co-authorship relations (source: DBLP), MARIOH achieved a reconstruction accuracy of over 98%, significantly outperforming existing methods, which reached only about 86%. Furthermore, leveraging the reconstructed higher-order structures led to improved performance in downstream tasks, including prediction and classification.
According to Kijung, “MARIOH moves beyond existing approaches that rely solely on simplified connection information, enabling precise analysis of the complex interconnections found in the real world.” Furthermore, “it has broad potential applications in fields such as social network analysis for group chats or collaborative networks, life sciences for studying protein complexes or gene interactions, and neuroscience for tracking simultaneous activity across multiple brain regions.”
The research was conducted by Kyuhan Lee (Integrated M.S.–Ph.D. program at the Kim Jaechul Graduate School of AI at KAIST; currently a software engineer at GraphAI), Geon Lee (Integrated M.S.–Ph.D. program at KAIST), and Professor Kijung Shin. It was presented at the 41st IEEE International Conference on Data Engineering (IEEE ICDE), held in Hong Kong this past May.
※ Paper title: MARIOH: Multiplicity-Aware Hypergraph Reconstruction ※ DOI: https://doi.ieeecomputersociety.org/10.1109/ICDE65448.2025.00233
<Figure 2. An example of the process of recovering high-dimensional relationships using MARIOH technology>
This research was supported by the Institute of Information & Communications Technology Planning & Evaluation (IITP) through the project “EntireDB2AI: Foundational technologies and software for deep representation learning and prediction using complete relational databases,” as well as by the National Research Foundation of Korea through the project “Graph Foundation Model: Graph-based machine learning applicable across various modalities and domains.”
2025.08.05 View 3185 -
KAIST Enables On-Site Disease Diagnosis in Just 3 Minutes... Nanozyme Reaction Selectivity Improved 38-Fold
<(From Left) Professor Jinwoo Lee, Ph.D candidate Seonhye Park and Ph.D candidate Daeeun Choi from Chemical & Biomolecular Engineering>
To enable early diagnosis of acute illnesses and effective management of chronic conditions, point-of-care testing (POCT) technology—diagnostics conducted near the patient—is drawing global attention. The key to POCT lies in enzymes that recognize and react precisely with specific substances. However, traditional natural enzymes are expensive and unstable, and nanozymes (enzyme-mimicking catalysts) have suffered from low reaction selectivity. Now, a Korean research team has developed a high-sensitivity sensor platform that achieves 38 times higher selectivity than existing nanozymes and allows disease diagnostics visible to the naked eye within just 3 minutes.
On the 28th, KAIST (President Kwang Hyung Lee) announced that Professor Jinwoo Lee’s research team from the Department of Chemical & Biomolecular Engineering, in collaboration with teams led by Professor Jeong Woo Han at Seoul National University and Professor Moon Il Kim at Gachon University, has developed a new single-atom catalyst that selectively performs only peroxidase-like reactions while maintaining high reaction efficiency.
Using bodily fluids such as blood, urine, or saliva, this diagnostic platform enables test results to be read within minutes even outside hospital settings—greatly improving medical accessibility and ensuring timely treatment. The key lies in the visual detection of biomarkers (disease indicators) through color changes triggered by enzyme reactions. However, natural enzymes are expensive and easily degraded in diagnostic environments, limiting their storage and distribution.
To address this, inorganic nanozyme materials have been developed as substitutes. Yet, they typically lack selectivity—when hydrogen peroxide is used as a substrate, the same catalyst triggers both peroxidase-like reactions (which cause color change) and catalase-like reactions (which remove the substrate), reducing diagnostic signal accuracy.
To control catalyst selectivity at the atomic level, the researchers used an innovative structural design: attaching chlorine (Cl) ligands in a three-dimensional configuration to the central ruthenium (Ru) atom to fine-tune its chemical properties. This enabled them to isolate only the desired diagnostic signal.
<Figure1. The catalyst in this study (ruthenium single-atom catalyst) exhibits peroxidase-like activity with selectivity akin to natural enzymes through three-dimensional directional ligand coordination. Due to the absence of competing catalase activity, selective peroxidase-like reactions proceed under biomimetic conditions. In contrast, conventional single-atom catalysts with active sites arranged on planar surfaces exhibit dual functionality depending on pH. Under neutral conditions, their catalase activity leads to hydrogen peroxide depletion, hindering accurate detection. The catalyst in this study eliminates such interference, enabling direct detection of biomarkers through coupled reactions with oxidases without the need for cumbersome steps like buffer replacement. The ability to simultaneously detect multiple target substances under biomimetic conditions demonstrates the practicality of ruthenium single-atom catalysts for on-site diagnostics>
Experimental results showed that the new catalyst achieved over 38-fold improvement in selectivity compared to existing nanozymes, with significantly increased sensitivity and speed in detecting hydrogen peroxide. Even in near-physiological conditions (pH 6.0), the catalyst maintained its performance, proving its applicability in real-world diagnostics.
By incorporating the catalyst and oxidase into a paper-based sensor, the team created a system that could simultaneously detect four key biomarkers related to health: glucose, lactate, cholesterol, and choline—all with a simple color change.
This platform is broadly applicable across various disease diagnostics and can deliver results within 3 minutes without complex instruments or pH adjustments. The findings show that diagnostic performance can be dramatically improved without changing the platform itself, but rather by engineering the catalyst structure.
<Figure 2.(a) Schematic diagram of the paper sensor (Zone 1: glucose oxidase immobilized; Zone 2: lactate oxidase immobilized; Zone 3: choline oxidase immobilized; Zone 4: cholesterol oxidase immobilized; Zone 5: no oxidase enzyme). (b) Single biomarker (single disease indicator) detection using the ruthenium single‑atom catalyst–based paper sensor.(c) Multiple biomarker (multiple disease indicator) detection using the ruthenium single‑atom catalyst–based paper sensor>
Professor Jinwoo Lee of KAIST commented, “This study is significant in that it simultaneously achieves enzyme-level selectivity and reactivity by structurally designing single-atom catalysts.” He added that “the structure–function-based catalyst design strategy can be extended to the development of various metal-based catalysts and other reaction domains where selectivity is critical.”
Seonhye Park and Daeeun Choi, both Ph.D. candidates at KAIST, are co-first authors. The research was published on July 6, 2025, in the prestigious journal Advanced Materials
-Title: Breaking the Selectivity Barrier of Single-Atom Nanozymes Through Out-of-Plane Ligand Coordinatio
- Authors: Seonhye Park (KAIST, co–first author), Daeeun Choi (KAIST, co–first author), Kyu In Shim (SNU, co–first author), Phuong Thy Nguyen (Gachon Univ., co–first author), Seongbeen Kim (KAIST), Seung Yeop Yi (KAIST), Moon Il Kim (Gachon Univ., corresponding author), Jeong Woo Han (SNU, corresponding author), Jinwoo Lee (KAIST, corresponding author
-DOI: https://doi.org/10.1002/adma.202506480
This research was supported by the Ministry of Science and ICT and the National Research Foundation of Korea (NRF).
2025.07.29 View 3831 -
Approaches to Human-Robot Interaction Using Biosignals
<(From left) Dr. Hwa-young Jeong, Professor Kyung-seo Park, Dr. Yoon-tae Jeong, Dr. Ji-hoon Seo, Professor Min-kyu Je, Professor Jung Kim >
A joint research team led by Professor Jung Kim of KAIST Department of Mechanical Engineering and Professor Min-kyu Je of the Department of Electrical and Electronic Engineering recently published a review paper on the latest trends and advancements in intuitive Human-Robot Interaction (HRI) using bio-potential and bio-impedance in the internationally renowned academic journal 'Nature Reviews Electrical Engineering'.
This review paper is the result of a collaborative effort by Dr. Kyung-seo Park (DGIST, co-first author), Dr. Hwa-young Jeong (EPFL, co-first author), Dr. Yoon-tae Jeong (IMEC), and Dr. Ji-hoon Seo (UCSD), all doctoral graduates from the two laboratories. Nature Reviews Electrical Engineering is a review specialized journal in the field of electrical, electronic, and artificial intelligence technology, newly launched by Nature Publishing Group last year. It is known to invite world-renowned scholars in the field through strict selection criteria. Professor Jung Kim's research team's paper, titled "Using bio-potential and bio-impedance for intuitive human-robot interaction," was published on July 18, 2025. (DOI: https://doi.org/10.1038/s44287-025-00191-5)
This review paper explains how biosignals can be used to quickly and accurately detect movement intentions and introduces advancements in movement prediction technology based on neural signals and muscle activity. It also focuses on the crucial role of integrated circuits (ICs) in maximizing low-noise performance and energy efficiency in biosignal sensing, covering thelatest development trends in low-noise, low-power designs for accurately measuring bio-potential and impedance signals.
The review emphasizes the importance of hybrid and multi-modal sensing approaches, presenting the possibility of building robust, intuitive, and scalable HRI systems. The research team stressed that collaboration between sensor and IC design fields is essential for the practical application of biosignal-based HRI systems and stated that interdisciplinary collaboration will play a significant role in the development of next-generation HRI technology. Dr. Hwa-young Jeong, a co-first author of the paper, presented the potential of bio-potential and impedance signals to make human-robot interaction more intuitive and efficient, predicting that it will make significant contributions to the development of HRI technologies such as rehabilitation robots and robotic prostheses using biosignals in the future. This research was supported by several research projects, including the Human Plus Project of the National Research Foundation of Korea.
2025.07.24 View 5504 -
KAIST Develops Customized Tactile Sensor That Can Detect Light Breath, Pressure and Sound
< Photo 1. (From left) Professor Inkyu Park of KAIST Department of Mechanical Engineering (ME), Dr. Jungrak Choi of ETRI, Ph.D. Candidate Donho Lee and M.S. Graduate Chankyu Han of KAIST ME >
When a robot grabs an object or a medical device detects a pulse, the tactile sensor is the technology that senses pressure like a fingertip. Existing sensors had disadvantages, such as slow responses or declining accuracy after repeated use, but Korean researchers have succeeded in developing a sensor that can quickly and accurately detect even light breath, pressure, and sound. This sensor can be used across a broad range — from everyday movements to medical diagnostics.
KAIST (represented by President Kwang Hyung Lee) announced on the 23rd of June that Professor Inkyu Park’s team from the Department of Mechanical Engineering, through a collaborative research project with the Electronics and Telecommunications Research Institute (ETRI, President Seung Chan Bang ) under the National Research Council of Science & Technology (NST, Chairman Young Sik Kim), has developed an innovative technology that overcomes the structural limitations of existing tactile sensors.
The core of this joint research is the implementation of a customized tactile sensor that simultaneously achieves flexibility, precision, and repeatable durability by applying Thermoformed 3D Electronics (T3DE).
< Figure 1. Comparative evaluation of soft elastomer–based 3D structure versus thermoforming-based 3D structure in terms of mechanical properties. >
In particular, soft elastomer-based sensors (rubber, silicone, etc. — materials that stretch and return to their original shape) have structural problems such as slow response times, high hysteresis*, and creep**, but this new platform operates precisely in diverse environments and overcomes these limitations.
*Hysteresis: A phenomenon where the previously applied force or change is retained like a “memory,” so that the same stimulus does not always produce the same result.
**Creep: The phenomenon where a material slowly deforms when a force is continuously applied.
T3DE sensors are manufactured by precisely forming electrodes on a 2D film, then thermoforming them into a 3D structure under heat and pressure. Specifically, the top electrodes and supporting pillar structures of the sensor are designed to allow the fine-tuning of the mechanical properties for different purposes. By adjusting microstructural parameters — such as the thickness, length, and number of support pillars — the sensor’s Young’s modulus* can be tuned across a broad range of 10 Pa to 1 MPa. This matches the stiffness of biological tissues like skin, muscle, and tendons, making them highly suitable as bio-interface sensors.
*Young’s modulus: An index representing a material's stiffness; this research can control this index to match various biological tissues.
The newly developed T3DE sensor uses air as a dielectric material to reduce power consumption and demonstrates outstanding performance in sensitivity, response time, thermal stability, and repeatable accuracy.
Experimental results showed that the sensor achieved △sensitivity of 5,884 kPa⁻¹, △response time of 0.1 ms (less than one-thousandth of a second), △hysteresis of less than 0.5%, and maintained a repeatable precision of 99.9% or higher even after 5,000 repeated measurements.
< Figure 2. Graphic Overview of thermoformed 3D electronics (T3DE) >
The research team also constructed a high-resolution 40×70 array, comprising a total of 2,800 densely packed sensors, to visualize the pressure distribution on the sole of the foot in real time during exercise and confirmed the possibility of using the sensor for wrist pulse measurement to assess vascular health. Furthermore, successful results were also achieved in sound-detection experiments at a level comparable to commercial acoustic sensors. In short, the sensor can precisely and quickly measure foot pressure, pulse, and sound, allowing it to be applied in areas such as sports, health, and sound sensing.
The T3DE technology was also applied to an augmented-reality(AR)-based surgical training system. By adjusting the stiffness of each sensor element to match that of biological tissues, the system provided real-time visual and tactile feedback according to the pressure applied during surgical incisions. It also offered real-time warnings if an incision was too deep or approached a risky area, making it a promising technology for enhancing immersion and accuracy in medical training.
KAIST Professor Inkyu Park stated, “Because this sensor can be precisely tuned from the design stage and operates reliably across diverse environments, it can be used not only in everyday life, but also in a variety of fields such as healthcare, rehabilitation, and virtual reality.”
The research was co-led as first authors by Dr. Jungrak Choi of ETRI, KAIST master’s student Chankyu Han, and Ph.D. candidate Donho Lee, under the overall guidance of Professor Inkyu Park. The research results were published in the May 2025 issue of ‘Science Advances’ and introduced to the global research community through the journal’s official SNS channels (Facebook, Twitter).
※ Thesis Title: Thermoforming 2D films into 3D electronics for high-performance, customizable tactile sensing
※ DOI: 10.1126/sciadv.adv0057
< Figure 3. The introduction of the study on the official SNS posting by Science Advances >
This research was supported by the Ministry of Trade, Industry and Energy, the National Research Foundation of Korea, and the Korea Institute for Advancement of Technology.
2025.06.23 View 7776