Highlights
Clear imaging deep inside the living brain has traditionally required advanced optical systems and precise correction techniques. A research team from our department has developed a physics-based AI computational algorithm that can restore blurred biological microscopy images more clearly without additional wavefront measurement hardware.
Professor Iksung Kang (School of Electrical Engineering), in collaboration with Professor Na Ji’s research team at UC Berkeley, has developed a technology that accurately corrects image aberrations in microscopes used for live biological imaging. The experimental design and algorithm development – the core components of this work – were led by Professor Kang during his postdoctoral fellowship in Professor Na Ji’s group. This method uses neural fields, a neural network-based approach that represents 3D spatial structures continuously to reconstruct clearer images and volumetric information.
The research team utilized two-photon fluorescence microscopy, a key technique for observing deep inside living biological tissue. This method generates fluorescence when two photons are absorbed nearly simultaneously, enabling localized imaging within biological samples. However, as light passes through thick tissue, differences in refractive index distort the optical wavefront, causing the image to become blurred, much like how objects appear distorted underwater. This phenomenon is known as optical aberration, in which wavefront distortions degrade the focus and clarity of an image.
Previously, correcting these distortions required additional complex and costly hardware, such as wavefront sensors, which measure how much the optical wavefront is distorted.
In contrast, the research team developed an algorithm that inversely calculates how light was distorted using only the acquired image data and corrects the distortion computationally. In other words, rather than simply sharpening a blurred image, the method incorporates the physical process of image formation to restore clearer images without additional wavefront measurement hardware.
The core of this technology is a machine learning algorithm based on neural fields. This algorithm models the distortion process that occurs as light propagates through biological tissue and the microscope system, enabling an integrated framework that simultaneously corrects optical aberrations caused by biological tissue, subtle motion of the living specimen, and mechanical alignment errors in the microscope.
As a result, the team demonstrated that clearer, higher-contrast images can be obtained from deep biological tissues without separate optical wavefront measurement or correction devices.
This research is particularly significant because it moves beyond the conventional approach that better imaging often requires more complex and expensive hardware. Instead, it shows that software-based computational algorithms can improve microscopy image quality. This approach is expected to help reduce the burden of research equipment and experimental procedures, and to support more precise biological imaging for a broader range of researchers.
Professor Iksung Kang stated, “This research shows the potential of combining optics and artificial intelligence to more accurately observe the inside of living biological systems. Moving forward, we plan to develop this into an intelligent optical imaging system where the microscope can identify the optimal imaging conditions.”
This study was published on April 13th in Nature Methods, a leading methodology journal in the life sciences.
※ Paper Title: Adaptive optical correction for in vivo two-photon fluorescence microscopy with neural fields
※ Authors: Iksung Kang (KAIST, Co-corresponding & First Author), Hyeonggeon Kim, Ryan Natan, Qinrong Zhang, Stella X. Yu, & Na Ji (UC Berkeley, Co-corresponding Author)
What if an AI, when asked about a minister appointed last month, returned the name of a predecessor from a year ago? This example illustrates a critical limitation of current AI systems: their inability to reliably reflect up-to-date information. Our university’s research team has developed a new evaluation framework that automatically incorporates changes in real-world information and detects “temporal errors” that can appear plausible on the surface. The study is expected to enhance AI reliability by providing a systematic benchmarking framework.
A research team led by Professor Steven Euijong Whang from the School of Electrical Engineering, in joint research with Microsoft Research, has developed a system that automatically evaluates and diagnoses the temporal reasoning capabilities of Large Language Models (LLMs) using temporal database technology.
For AI to earn users’ trust, it must be able to accurately understand real-world information that changes over time. However, existing evaluation methods have largely focused on whether answers are simply right or wrong, or have examined only a narrow set of temporal relations, making them insufficient for evaluating the wide range of question scenarios that arise in real-world environments.
To overcome this challenge, the research team integrated “Temporal Database” design theory—an approach refined and validated over the past 40 years—into AI evaluation for the first time. By leveraging the temporal dependencies and relational structure of data, the technology can automatically generate 13 types of complex time-sensitive questions directly from the database, eliminating the need for researchers to manually create evaluation questions.
In particular, this technology marks a major innovation by replacing the conventional approach of manually writing evaluation questions with a data-driven method that generates them automatically. By automating the entire process—from question generation to answer derivation and verification—based on the database, it also reduces maintenance burden by eliminating the need to manually revise evaluation items.
When real-world information changes, the evaluation questions, answers, and verification criteria are automatically updated simply by revising the relevant data in the database. Although the latest information must still be provided by external data sources or administrators, the framework is designed to automatically conduct the evaluation once the data has been updated.
Additionally, going beyond traditional methods that assess only whether a final answer is correct or incorrect, the research team introduced a new metric that evaluates the factual validity of the dates or time periods used during the answering process. Using this metric, the team achieved a 21.7% improvement in detecting “Temporal Hallucinations”—cases in which an answer appears correct on the surface but is based on faulty temporal reasoning—compared with previous methods.
The database-based approach also improved evaluation efficiency. By eliminating the reliance on unnecessary data, the research team reduced the amount of input data required by an average of 51% compared with previous methods and demonstrated its effectiveness in reducing evaluation maintenance costs.
Professor Steven Euijong Whang stated, “This research shows that classical database design theory can play a crucial role in addressing the reliability challenges of today’s AI systems. By transforming large amounts of domain-specific data into evaluation resources, we expect this work to provide a practical foundation for verifying AI performance in various fields such as medicine and law.”
Soyeon Kim, a PhD student at KAIST, participated as the lead author of this study, and Jindong Wang (Microsoft Research, currently at William & Mary) and Xing Xie (Microsoft Research) participated as co-authors. The research results will be presented this April at ICLR 2026, the most prestigious academic conference in the field of artificial intelligence.
※ Paper Title: Harnessing Temporal Databases for Systematic Evaluation of Factual Time-Sensitive Question-Answering in Large Language Models
※ Paper Link: https://arxiv.org/abs/2508.02045
Meanwhile, this research was conducted with support from Microsoft Research, the National Research Foundation of Korea, and the Institute for Information & Communications Technology Planning & Evaluation (IITP) Global AI Frontier Lab projects (RS-2024-00469482, RS-2024-00509258).
A new technology has been developed that allows light to be “designed” into desired forms, potentially making Artificial Intelligence (AI) and communication technologies faster and more accurate. A KAIST EE research team has developed an “integrated photonic resonator”—a core component of next-generation optical integrated circuits that process data using light. The research is particularly significant as it was led by an undergraduate student. This technology is expected to serve as a key foundation for next-generation security technologies such as high-speed data processing and quantum communication.
A research team led by Professor Sangsik Kim from the School of Electrical Engineering, in collaboration with Professor Jae Woong Yoon’s team from the Department of Physics at Hanyang University (President Kigeong Lee), has developed a new integrated photonic resonator structure capable of freely controlling optical signals by utilizing light interference (the phenomenon where two light waves meet and influence each other).
Photonic Integrated Circuits (PICs) process data at ultra-high speeds and with low power consumption using light. They are garnering significant attention as a fundamental platform technology for next-generation fields such as AI, data centers, and quantum information processing.
The core of this technology lies in the precision with which light can be controlled. Specifically, the ability to freely adjust the spectrum (color or wavelength distribution) and phase response (timing or wave position) of optical signals is essential for implementing high-performance optical communication and computing. However, conventional methods have faced fundamental limitations.
The integrated photonic resonator (optical resonator) focused on by the research team is a key optical device that traps light in a specific space to amplify it or select specific colors (wavelengths), similar to how the body of a musical instrument amplifies sound. However, existing single-bus resonators have had limitations in precisely adjusting the phase and spectrum of optical signals.
To overcome these challenges, the research team introduced a “dual-bus” structure. This design allows light that has passed through the resonator to recombine with light that has not, enabling precise control over interference. This allows for the free design of optical signals into desired forms, making it possible to control various types of light signals that were previously difficult to implement.
By applying this technology, the research team secured new characteristics for more precise control of wavelength properties and presented new possibilities for non-linear frequency conversion research (changing the color of light). Utilizing this technology enables faster and more accurate data processing, which is expected to provide the groundwork for performance enhancements in future high-speed data centers, AI accelerators, and quantum communication systems.
This research is especially meaningful as it was led by an undergraduate student. Taewon Kim, an undergraduate student who conducted the study through the KAIST Undergraduate Research Program (URP), stated, “I was able to develop the resonator principles I learned in the Introduction to Integrated Optics class into actual device designs and a published paper.”
Professor Sangsik Kim remarked, “This study goes beyond proposing a new device; it demonstrates that by precisely analyzing previously overlooked optical characteristics, physical limitations can be overcome. We expect this to contribute broadly to the development of optics-based AI accelerators and optical communication technologies.”
KAIST undergraduate student Taewon Kim participated as the lead author of this study, and the results were published on March 6th in the international optics journal, Laser & Photonics Reviews.
※ Paper Title: Dual-bus resonator for multi-port spectral engineering
※ DOI: 10.1002/lpor.202502935 (Authors: Taewon Kim, Mehedi Hasan, Yu Sung Choi, Jae Woong Yoon, and Sangsik Kim)
This research was supported by the KAIST URP Program, the Institute of Information & Communications Technology Planning & Evaluation (IITP), the U.S. Asian Office of Aerospace Research and Development (AOARD), and the National Research Foundation of Korea (NRF).
Instead of applying ointment and attaching a bandage, a ‘smart patch that regulates treatment intensity on its own just by being attached’ has appeared. Our university’s research team has developed a ‘self-regulating OLED wound healing patch’ that combines light and drugs to pull up the wound recovery speed by about twice. It is expected to develop into an intelligent treatment technology where light regulates drug release according to the patient’s condition in the future.
A research team led by Professor Kyung Cheol Choi of the School of Electrical Engineering, together with Dr. Daekyung Sung of the Korea Institute of Ceramic Engineering and Technology (President Jong-seok Yoon) and Professor Chan-Su Park’s team at Chungbuk National University (Acting President Yu-sik Park), developed a ‘self-regulating wound healing patch’ technology that combines Organic Light Emitting Diodes (OLED) and a Drug Delivery System (DDS).
Ointments can cause side effects when overused, and Photobiomodulation (PBM)* treatment, which helps cell regeneration using light, also had limitations in that its effect decreased if the appropriate amount was exceeded.
*PBM (Photobiomodulation): A non-invasive treatment method that promotes the recovery of cells and tissues using low-intensity light.
The research team focused on solving the limitations of existing treatment methods, which make it difficult to appropriately regulate treatment intensity. The core of this research is that ‘light regulates the medicine.’ When light is applied, Reactive Oxygen Species (ROS) are generated in the body, and this substance plays a role in stimulating nanoparticles so that drugs are released.
In other words, the amount of reactive oxygen species generated varies according to the intensity of light, and the amount of drug release is naturally regulated accordingly. When light is applied, cell regeneration is promoted, and at the same time, the ROS generated at this time acts as a ‘switch’ so that the drug is automatically released only as much as necessary. It is an ‘intelligent treatment method’ in which the treatment maintains its optimal level on its own even if a person does not regulate it separately. Simply put, it is a ‘self-regulating treatment patch’ where the medicine automatically comes out in an appropriate amount according to the intensity of the light when it is shone.
The research team produced a 630-nanometer (nm) wavelength OLED patch that closely adheres to the skin. This patch was designed to deliver light evenly to induce cell regeneration while releasing an appropriate amount of antioxidant drugs, such as Centella asiatica (commonly known as tiger grass) extract, a plant-derived ingredient well known for its skin regeneration effects.
In addition, it was produced in a wearable form that perfectly adheres to the curves of the skin to reduce light energy loss, and it maintains a temperature of about 31 degrees Celsius even during long-term use, allowing it to be used safely without the risk of low-temperature burns. Stability, maintaining performance for more than 400 hours, was also confirmed, securing the possibility of application to actual medical devices.
The effect was confirmed through experiments. In skin cell experiments, ‘combined treatment’ using light and drugs together showed faster recovery than single treatment. In mouse experiments, the wound recovery rate was 67% as of the 14th day of treatment, recording a healing speed about twice as fast as that of the control group (35%). The quality of healing was also significantly improved, such as skin thickness and barrier protein formation recovering to normal levels.
Professor Kyung Cheol Choi stated, “This research is an example of expanding OLED-based light treatment beyond the level of simply applying it to the role of regulating the treatment, and into a combined treatment platform where drug release is automatically regulated according to the wound status. We plan to develop it into an intelligent treatment technology that can be applied to various wounds and diseases and reacts on its own according to the patient’s body condition.”
In this research, Hyejeong Yeon, a doctoral student at the KAIST School of Electrical Engineering, participated as the first author. It was published online in the international academic journal ‘Materials Horizons’ last January and was selected as the Front Cover Paper in March.
※ Paper title: A self-regulating wearable OLED patch for accelerated wound healing via photobiomodulation-triggered drug delivery
※ DOI: https://doi.org/10.1039/D5MH02129D (Authors: Hyejeong Yeon, Sohyeon Yu, Minhyeok Lee, Sangwoo Kim, Yongjin Park, Hye-Ryung Choi, Won Il Choi, Chang-Hun Huh, Yongmin Jeon, Chan-Su Park, Daekyung Sung, and Kyung Cheol Choi)
This research was conducted with the support of the Future Discovery Convergence Science and Technology Development Program (2021M3C1C3097646) carried out through the National Research Foundation of Korea (NRF) of the Ministry of Science and ICT.
A EE research team has developed quadrupedal robot technology that not only enables walking by estimating terrain without visual information, but also allows the robot to perceive its surroundings through cameras and LiDAR sensors and make its own decisions while walking, much like animals that visually examine terrain and adjust their steps. This technology is also expected to be extended to various robotic platforms such as wheeled-legged robots and humanoid robots.
A research team led by Professor Hyun Myung from the School of Electrical Engineering, in collaboration with the lab’s startup EuRoboTics Co., Ltd., has developed “DreamWaQ++,” a quadrupedal robot control technology that recognizes terrain based on visual information and adjusts locomotion strategies in real time.
The previously developed “DreamWaQ” by this research team is a “blind locomotion” technology that estimates terrain using only proprioceptive sensing such as joint encoders and inertial sensors, enabling robust movement even without visual information. It allows stable walking even in environments where visual information is difficult to obtain, such as disaster situations, but has the limitation that the robot can only adjust its movement after its legs directly contact obstacles.
The newly developed DreamWaQ++ overcomes this limitation by combining proprioceptive sensing with exteroceptive sensing based on cameras and LiDAR. The key is that it enables “perception-based locomotion,” in which the robot recognizes obstacles in advance and proactively adjusts its walking strategy, going beyond simple reactive control to understanding and making decisions about the environment.
To achieve this, the research team designed a multimodal reinforcement learning architecture and implemented it to enable real-time control based on lightweight computation. In addition, it simultaneously secured stability by automatically switching to locomotion based on other sensory modalities when sensor errors occur, and scalability that allows application to various robotic platforms.
Performance was also demonstrated through experiments. The robot equipped with DreamWaQ++ showed performance surpassing existing technologies in various challenging environments.
In stair locomotion experiments, it completed a course of 50 steps (30.03 m horizontally, 7.38 m vertically) in just 35 seconds, outperforming both blind locomotion controllers and commercial perception-based controllers.
In steep slope environments, it stably climbed a 35° incline, which is 3.5 times steeper than the training condition (10°), and actively adjusted its posture to reduce the rear leg motor torque by approximately 1.5 times compared to existing methods.
In addition, in various obstacle scenarios, it demonstrated learning-based perception capability by autonomously selecting more efficient paths without separate path planning, and in uncertain drop terrains, it exhibited “exploration behavior,” where it voluntarily stops to inspect the ground before moving.
Along with this, it demonstrated high agility by overcoming obstacles of 41 cm, exceeding the robot’s height, even while carrying a payload of 2.5 kg. In simulation, it was shown that it can handle obstacles up to 1.0 m with ANYmal-C (a representative quadrupedal robot developed at ETH Zurich) and up to 1.5 m with KAIST HOUND (a quadrupedal robot developed by Professor Hae-Won Park’s group at KAIST).
In particular, even though it was trained only on relatively low obstacles (27 cm), it achieved a success rate of about 80% on actual higher stairs of 42 cm. This means that the robot is not simply repeating learned situations but has the ability to adapt to new environments on its own.
The research team expects that this technology can be applied in environments where conventional wheeled robots have difficulty accessing, such as disaster response, industrial facility inspection, forestry, and agriculture.
Professor Hyun Myung said, “This research shows that robots have advanced beyond simply moving to a level where they understand the environment and make decisions on their own,” adding, “We will further expand this into intelligent mobility technologies applicable in various real-world environments.”
This study was led by I Made Aswin Nahrendra (first author, current researcher at Krafton, KAIST PhD graduate), with co-authors Byeongho Yu (EuRoboTics Co., Ltd. CEO), Minho Oh (EuRoboTics Co., Ltd. CTO), Dongkyu Lee (EuRoboTics Co., Ltd. CTO), Seunghyun Lee (KAIST), Hyeonwoo Lee (KAIST), and Dr. Hyungtae Lim (MIT postdoctoral researcher). The study was published in February in the world-renowned robotics journal IEEE Transactions on Robotics (T-RO).
※ Paper title: DreamWaQ++: Obstacle-Aware Quadrupedal Locomotion With Resilient Multi-Modal Reinforcement Learning
※ link to original Paper: https://arxiv.org/abs/2409.19709
※ Videos demonstrating DreamWaQ++ operation and locomotion
- Main video
- Additional video
- Humanoid application video of improved DreamWaQ1
- Humanoid application video of improved DreamWaQ2
- Wheeled-legged robot application video of improved DreamWaQ
- Project page
This research was supported by the Korea Evaluation Institute of Industrial Technology (KEIT) under the Ministry of Trade, Industry and Energy (Project No. 20018216, “Development and Field Deployment of Mobility Intelligence Software for Autonomous Locomotion of Walking Robots in Dynamic and Unstructured Environments”), and by the Korea Forest Service (Korea Forestry Promotion Institute) through the Forest Science and Technology R&D Program (Project No. RS-2025-25424472).
Professor Choi is an internationally recognized scholar in wireless networking and mobile systems/computing. An IEEE Fellow recognized for his significant contributions to the field, he has held distinguished leadership roles in both academia and industry, including serving as a professor at Seoul National University and, most recently, as an Executive Vice President at Samsung Electronics.
His research has advanced intelligent network optimization, digital twins, low-latency mobile systems, and indoor localization. At KAIST, his work will focus on AI-native network architectures, autonomous network operations using LLMs and multi-agent systems, and next-generation infrastructure for physical AI.
Professor Choi’s office is located in N1 Room 616. For more details on his research, please refer to the website below.
► Visit Professor Sunghyun Choi’s Homepage(Click)
|
Education
Professional Experience
Major Publications
Vision
Research Plan
Assigned Course
|
