%41%49
-
KAIST Develops AI to Detect ‘Foreign-Linked Opinion Manipulation’ in 110 Million News Comments
During election seasons or major national issues, online news comment sections often become heated spaces of conflict across gender, generation, and political lines. For years, there have been persistent concerns that behind some of these conflicts may lie “foreign winds,” or interventions by foreign actors seeking to manipulate public opinion and deepen social divisions. A KAIST research team has now developed a technology that uses big data on two decades worth of news comments and artificial intelligence (AI) to precisely detect traces of such hidden influence operations.
KAIST (President Choongsik Bae) announced on the 12th of August that a joint research team led by Professor Wonjae Lee of the Graduate School of Culture Technology, Professor Meeyoung Cha of the School of Computing (Scientific director at Max Planck Institute for Security and Privacy) and Professor Alice Oh of the School of Computing, in collaboration with Professor Thorsten Holz of the Max Planck Institute, has developed an explainable AI technology that automatically detects patterns suspected of foreign-linked influence operations in online news comments and provides specific evidence for its judgments.
The organized and repeated posting of comments or content by certain actors to shape public opinion in a desired direction is known as an “online influence operation.” Existing AI-based detection technologies have had a key limitation: even when they classify certain accounts as belonging to a “blacklist,” they often fail to provide clear evidence explaining why those accounts should be considered influence-operation accounts.
To overcome this limitation, the research team used 70 foreign-linked accounts previously identified by the Institute for National Security Strategy as starting points. They then tracked groups of accounts connected to them or repeatedly commenting on the same news articles, ultimately collecting and analyzing a large-scale dataset of 110 million comments posted on Naver News over a 20-year period from 2006 to 2025.
In particular, the AI developed by the research team examines accounts through a careful three-step process. First, it checks whether there are clues suggesting that the author may be linked to a foreign source. Second, it examines whether the comment contains emotionally polarizing expressions, such as moral condemnation or blind praise. Third, it identifies which country or target the emotional framing is directed toward.
The model does not stop at simply labeling an account as suspicious. It also highlights the specific phrases in the comments that served as the basis for its judgment. The team further combined this with multidimensional behavioral-pattern analysis, including account activity frequency, account lifespan, and activity links with other suspected accounts. As a result, among approximately 4 million Naver News users, the model ultimately identified 23,998 accounts exhibiting patterns consistent with suspected public-opinion manipulation.
The analysis also revealed the more subtle strategy of these suspected accounts. Their main target was not the victory of a particular political camp, but rather the maximization of division and confrontation within Korean society.
Among the top 10 targets that drew the highest public engagement, measured through likes and other reactions, seven were prominent domestic political figures. Notably, the attacks were not concentrated on a single party or ideology. Former and current presidents, presidential candidates, and political parties from both progressive and conservative camps were targeted across the spectrum. According to the research team, this suggests a sophisticated strategy aimed not so much at supporting a particular group, but at inflaming domestic political conflict and increasing social distrust and polarization.
This study is significant because it provides data-based evidence for influence-operation activity that had previously been discussed largely in terms of suspicion, while also offering a potential defense mechanism for protecting healthy online public discourse. In the future, portal platforms and related organizations could use this technology during elections or national crises to monitor the influx of suspicious accounts in real time and prioritize the review of coordinated attacks against domestic political figures. However, the research team emphasized that the AI should not be used to block accounts indiscriminately, but rather as an explainable content-moderation tool that supports the judgment of expert reviewers.
Professor Wonjae Lee said, “By analyzing 20 years of data, we found that suspected accounts tended to use messages criticizing Korea and domestic political figures rather than directly praising foreign countries, and that these messages gained higher visibility,” adding, “This research can provide empirical criteria for when and which messages platforms and monitoring organizations should prioritize for review, especially during socially sensitive periods such as elections.”
Professor Alice Oh said, “This is a meaningful achievement in which AI precisely identified not only the surface meaning of words in massive comment datasets, but also subtle emotional patterns and organized behavioral signals intended to provoke conflict,” adding, “It can become a powerful defense system against online influence operations, which are becoming increasingly sophisticated.”
Professor Meeyoung Cha said, “This study goes beyond simple blacklist-account analysis and represents the outcome of actionable data science that addresses real-world problems and drives practical change,” adding, “In an online environment where social conflict is deepening, we hope this technology will serve as a practical tool for protecting the transparency and trustworthiness of the digital public sphere.”
This research was led by KAIST Ph.D. candidate Jaehong Kim and master’s student Hyeonseung Kim as co-first authors. The paper is scheduled to be presented at the USENIX Security Symposium 2026, one of the most prestigious conferences in the field of computer security.
Paper title: Cross-National Information Attacks: A Two-Decade Analysis of Troll Behavior in Korea,
DOI: 10.48550/arXiv.2606.22785
This research was supported by the Hyundai Motor Chung Mong-Koo Foundation, the Institute of Information & Communications Technology Planning & Evaluation, and the National Research Foundation of Korea, funded by the Ministry of Science and ICT.
2026.08.12 View 154 -
KAIST Held Inauguration Ceremony for 18th President Choongsik Bae, Unveiling Vision of "Fundamentals First, Innovation Forward"
KAIST announced that it held an inauguration ceremony for its 18th president, Choongsik Bae, at the KAIST Auditorium on Monday, August 10. At the ceremony, the university unveiled "Fundamentals First, Innovation Forward" as its new vision.
The ceremony officially presented President Bae's philosophy on university governance and his vision for KAIST's future to the KAIST community and the public. Departing from the conventional format of a formal inaugural address, President Bae personally explained his vision and the strategies for implementing it. Professor Yiyun Kang of the Department of Industrial Design directed the stage production, bringing KAIST's future vision to life through an intuitive and immersive presentation.
The event built on the innovation advanced under KAIST's 17th president, Kwang Hyung Lee, while introducing new leadership and development strategies that will guide the university toward its 60th anniversary. Distinguished guests from Korea and abroad attended, including Deputy Prime Minister and Minister of Science and ICT Kyung Hoon Bae, former KAIST President Kwang Hyung Lee, and ambassadors to Korea from key countries.
In his inaugural address, President Bae presented "Continuity & Innovation" as the central philosophy of his administration. He aimed to preserve the values KAIST has cultivated over the past 55 years -- Creativity, Challenge, and Caring -- while pursuing innovation across education, research, entrepreneurship, and administration in response to AI-driven transformation and intensifying global competition for technological leadership.
The new vision, "Fundamentals First, Innovation Forward," rests on two foundational principles: people strongly grounded in fundamental disciplines, humanistic insight, and AI capabilities; and an organization characterized by autonomy, accountability, and efficiency. On these foundations, KAIST aims to achieve world-class excellence in education, research, entrepreneurship, and internationalization.
To realize this vision, KAIST will pursue the following five development strategies, collectively called the Beyond Series:
Beyond AI – AI for Everyone: Create a leading environment for education and research that moves beyond today's AI toward Humanistic AI, Democratic AI, and Agentic AI.
Beyond Laboratory – Innovative Research and Entrepreneurship: Move beyond the laboratory to advance deep-tech innovation in partnership with industry and society and build a global startup ecosystem.
Beyond Barriers – An Efficient and Open University: Remove barriers so that members can devote themselves to research and education, underpinned by transparent governance and a culture and systems built on trust.
Beyond Carbon – Sustainability and a Greener Future: Strengthen research to address the climate crisis and create an environmentally responsible, carbon-neutral campus grounded in ESG and the UN Sustainable Development Goals.
Beyond KAIST – Toward the World and the Future through Global Connect: Connect global talent, universities, research institutions, companies, and local communities; foster a more international campus; expand international joint research; and strengthen global and regional partnerships.
KAIST plans to make AI not merely a technology for specific disciplines or specialists, but a common language and general-purpose tool across all fields. By strengthening foundational education and interdisciplinary AI education, KAIST aims to push beyond merely using AI effectively toward leading AI innovation.
KAIST will also expand research in physical AI, AI that operates in the real world, including robotics, autonomous driving, and advanced manufacturing, as well as in strategic technologies such as quantum science, climate technology, and energy technology. Building on world-class basic research, KAIST will expand industry collaboration, technology commercialization, and global entrepreneurship, creating a cycle in which research outcomes drive innovation in industry and society.
KAIST will expand the establishment of corporate satellite laboratories and collaborative research centers. It will also support joint research and development with companies by building AI Autonomous Labs that integrate AI into the R&D process, creating a new research environment in which AI designs and conducts experiments and analyzes the results. The university will introduce specialized entrepreneurship education for newly admitted students and establish a model that combines classroom instruction with hands-on training, involving alumni entrepreneurs and industry professionals.
The inauguration also featured case studies of KAIST alumni using AI to drive innovation in industry and research. Dr. Hyeon-Sook Yoon from Korea Shipbuilding & Offshore Engineering (KSOE) presented the use of digital twins in the shipbuilding and maritime industries, while Dr. Ji-Yong Shin from Samsung Electronics' Semiconductor R&D Center discussed the use of AI in semiconductor manufacturing. Professor Joonsik Hwang of KAIST then discussed the development and applications of physical AI in automobiles, mobility, robotics, and other fields.
KAIST plans to build an AI Native Campus that organically connects AI Interactive Education in education, AI Autonomous Labs in research, AI Agent Administration in administration, and AI Energy Convergence in infrastructure.
KAIST will also build an AI-based digital administration system to streamline or eliminate unnecessary regulations and procedures so that faculty and students can focus more fully on education and research. The campus will also become a living lab where climate and energy technologies are developed and validated, while global cooperation will be strengthened by recruiting outstanding international students and faculty, expanding international joint research, and broadening dual-degree programs.
In his address, President Bae said, “We will carry forward the proud tradition we have inherited: our vision of becoming a Global Value-Creative Leading University and our C-Cube core values of Creativity, Challenge, and Caring. Building on this foundation, we will pursue the innovation needed to move toward our new goal, Fundamentals First, Innovation Forward.” He added, “Grounded in strong fundamentals across both our people and our institution, we will advance five strategic priorities—AI, global entrepreneurship, a stronger focus on education and research, sustainable growth, and internationalization—and further establish KAIST as a world-leading university.”
He also emphasized, “I will listen with an open mind and act with determination. As both a facilitator and a servant leader, I will empower every member of the KAIST community to pursue their aspirations with confidence and fulfillment. Together, we will take KAIST beyond innovation—establishing it as a university that sets new standards and a national innovation platform shaping the future of science and technology in Korea.”
President Bae is an internationally recognized mechanical engineer and energy scientist specializing in carbon-neutral transportation power systems and sustainable mobility technologies. He earned his bachelor’s and master’s degrees in aerospace engineering from Seoul National University and a Ph.D. in mechanical engineering from Imperial College London. Since joining KAIST in 1998, he has served in leadership roles including Chair of the Department of Mechanical Engineering, Dean of the College of Engineering, and Director of the Mobile Clinic Module Project during the COVID-19 pandemic, gaining broad experience in education, research, and university administration.
He has also contributed to energy and carbon-neutrality research and to national science and technology policy as chair of the International Energy Agency's Technology Collaboration Programme on Sustainable Combustion, chair of the Climate Division of the Ministry of Foreign Affairs' Science and Technology Diplomacy Advisory Committee, and chair of the Society of Carbon-Neutral Fuel Technology. He was the first Korean researcher in the powertrain field to be elected an SAE Fellow and has received honors including a Presidential Commendation and a Merit Award from the National Assembly of the Republic of Korea.
KAIST presented the inauguration as a ceremony marking the start of a new presidency and as a forum for sharing the university's future vision and implementation strategies. The occasion marked KAIST's move beyond "a KAIST that embraces challenges" toward "a KAIST that sets the next standard for innovation," as it pursues its goal of becoming a world-leading university for innovation.
2026.08.10 View 230 -
KAIST Develops ‘Chameleon AI Semiconductor’ with Programmable Response Speeds
AI semiconductors are becoming more programmable. KAIST researchers have developed a device whose response characteristics can be programmed to process data changing at different speeds. The technology reduced prediction errors for time-varying data by up to 40-fold and is expected to enhance real-time AI performance in autonomous vehicles, robots, and wearable devices.
KAIST (President Choongsik Bae) announced on August 7 that a research team led by Chair Professor Shinhyun Choi from the School of Electrical Engineering and the Graduate School of Semiconductor Technology has developed a programmable dynamic memtransistor (PDM), a semiconductor device whose time-response characteristics can be adjusted to multiple states and retained, as well as an integrated array based on the device.
A memtransistor is a next-generation semiconductor device that combines the information-storage function of memory with the computing function of a transistor. In the developed PDM, the ability to process data while retaining previous information allows its response characteristics to be adjusted and retained for incoming data.
Today’s computers and smartphones require complex software processing to analyze data that changes over time, resulting in large computational loads and high power consumption. To address this, researchers have been studying technologies that allow semiconductor hardware itself to process data directly. However, conventional devices have had fixed response speeds that cannot be changed once the device is fabricated.
The research team overcame this limitation by introducing a dual-layer structure inside the transistor, combining a charge storage layer that accumulates and processes data with an electron trapping layer that controls the response speed in a nonvolatile manner.
In the PDM developed by the research team, incoming data is processed in the charge storage layer, while the electron trapping layer controls, across multiple levels, the recovery speed at which the semiconductor returns to its original state. In experiments, the team succeeded in tuning the current recovery time over an approximately 5-fold range and the characteristic frequency over a range of more than 10-fold.
In particular, in experiments involving the prediction of data in which fast and slow changes are intricately mixed, the PDM reduced prediction errors by as much as 40 times compared with conventional fixed-response semiconductor devices. The PDM enables accurate information processing even when handwriting or object-movement speeds vary, by using response characteristics configured to match different input timescales. Once the response characteristics are set, the device remembers them without requiring a continuous external power supply, and it does not require complex preprocessing of input data. Because it is fully compatible with materials used in widely adopted commercial semiconductor processes, it is also highly advantageous for mass production and commercialization.
The research team fabricated a PDM array and used it to predict complex data, confirming that it achieved accuracy comparable to conventional software-based systems while consuming far less energy.
“This study demonstrates an AI semiconductor whose response characteristics can be programmed to efficiently process data changing at different speeds,” said Chair Professor Choi. “We expect it to become a core technology that improves the performance of AI devices such as autonomous vehicles, robots, and wearables while reducing their power consumption.”
This research was led by KAIST Graduate School of Semiconductor Technology Ph.D. candidate Dae-won Kim as the first author, with Yoonho Cho, Seokho Seo, Yujin Kim, See-On Park, Taehwan Jang, and Chaebin Park participating as co-authors. Young Taek Oh and Fellow Jae-Duk Lee of Samsung Electronics’ Semiconductor R&D Center also participated as co-authors, and Chair Professor Shinhyun Choi served as the corresponding author. The research was published in July in the internationally renowned journal Nature Communications on July 4.
Paper title: Programmable memtransistor array with temporal dynamics modulation for efficient time-series data processing,
DOI: https://doi.org/10.1038/s41467-026-75211-5
This research was supported by the National R&D Program through the National Research Foundation of Korea funded by the Ministry of Science and ICT, the ETRI R&D Support Program of the Institute of Information & Communications Technology Planning & Evaluation, the HRD Program for Industrial Innovation of the Korea Institute for Advancement of Technology funded by the Ministry of Trade, Industry and Energy, Samsung Electronics, and others.
2026.08.07 View 513 -
KAIST brings ‘giant batteries’ closer to commercialization in the AI data center era
The explosive growth of AI data centers has brought the commercialization of "giant batteries" one step closer. A KAIST research team has developed a process that cuts the production time for a core material used in large-capacity batteries by 67%, resolving the largest production bottleneck standing in the way of commercialization.
KAIST (President Choongsik Bae) announced on August 5 that a research team led by Professor Hee-Tak Kim from the Department of Chemical and Biomolecular Engineering has developed a process for producing the core electrolyte of vanadium redox flow batteries (VRFBs)—a leading candidate for large-capacity energy storage systems (ESS)—faster and more stably.
As AI data centers operate around the clock in growing numbers, large-capacity ESS that can store electricity generated from solar and wind power and supply it reliably when needed have become increasingly important.
Because VRFBs use nonflammable, water-based electrolytes, they have a lower fire risk than many conventional battery systems. And their energy-storage capacity can be scaled by increasing the amount of electrolyte stored in external tanks. This has drawn attention to VRFBs as ultra-large batteries suited to AI data centers and renewable energy storage. However, producing the vanadium electrolyte with an average oxidation state of 3.5+—the standard starting composition for VRFB operation— has been slow and costly, making it a critical obstacle to commercialization.
The conventional process first produces the electrolyte through chemical reduction—a reaction in which a chemical reducing agent causes vanadium ions to gain electrons—and then refines it through electrochemical reduction, which applies electric current to adjust the vanadium ions' electron state to the desired level. This final electrochemical step, however, relies on a costly VRFB stack and significant electrical energy, increasing both operational complexity and capital costs.Beyond the limitations of the electrochemical reduction process, the research team found, for the first time, that the alternative chemical reduction process also suffers from a distinct kinetic bottleneck. The reaction rate slows sharply at a specific point, much like highway traffic suddenly backing up at a bottleneck. This bottleneck occurs when the average vanadium oxidation state reaches approximately +4.1, an intermediate stage in the production of V3.5+ electrolyte.
In previous research, the team had replaced the conventional electrochemical adjustment step with a Pt/C-catalyzed reduction process, preventing the waste of leftover electrolyte. In the present study, it further extended the catalytic process into the bottleneck region of oxalic-acid-based chemical reduction. By switching from chemical to catalytic reduction at an average oxidation state of approximately +4.1, the team was able to bypass the slowest stage of the production process.
As a result, production time for V3.5+ electrolyte was cut by 67% compared to the conventional process. The switch also eliminated residual oxalic acid, an impurity that can degrade battery performance. The same catalyst was reused more than 2,500 times without a notable drop in performance, demonstrating the process's viability for industrial-scale production.
"This study combined reaction engineering principles with thermodynamic predictions to identify the rate-determining step in the chemical reduction and redesigned the electrolyte production process to overcome this major bottleneck to the commercialization of large-scale batteries," said Hee-Tak Kim, professor in the Department of Chemical and Biomolecular Engineering. He added, "By scientifically identifying the conditions under which the catalyst operates stably without degrading in the electrolyte environment, we resolved a production bottleneck relevant to industry, and we expect this to significantly accelerate the commercialization of large-capacity energy storage technology."
Kyunghwa Seok, a PhD candidate in the Department of Chemical and Biomolecular Engineering, led the research as first author. The findings were published online in Advanced Energy Materials—a leading international journal in the energy field—on May 7. In particular, in recognition of its academic significance, the study was selected as the cover article for Issue 34, which is scheduled to be published online in early September.
Paper title: Streamlined V3.5+ Electrolyte Production by Leveraging Chemical and Catalytic Reductions
DOI: https://doi.org/10.1002/aenm.71029
Authors: Kyunghwa Seok (KAIST, first author), Minseong Kang (KAIST, second author), and Hee-Tak Kim (KAIST, corresponding author).
This research was supported by Lotte Chemical.
2026.08.05 View 509 -
KAIST and Seoul National University Students Hold 100-Hour Robot Hackathon to Nurture Physical AI Talent
KAIST (President Choongsik Bae) announced on August 4 that RoboticUS, a joint student organization formed by students from KAIST and Seoul National University, is holding the inaugural Robot Hackathon at KAIST from August 3 to 8.
"In the era of Physical AI, we need convergence talent who can go beyond building good AI to design and implement robots and systems that move the real world based on AI," said Choongsik Bae, President of KAIST. "This hackathon, planned and run entirely by students, is a good example of KAIST's culture of challenge and collaboration, and we expect it to become a new educational model for turning future technologies into reality," he added.
The hackathon puts this educational philosophy directly into students' hands. It is Korea's first student-led Physical AI robot hackathon, planned and run by students from KAIST and Seoul National University across institutional boundaries. Participants experience the entire process of designing and building working robots, developing hands-on capabilities that integrate AI and hardware.
The event is hosted by RoboticUS, a nonprofit student organization formed jointly by MR, a robotics club in KAIST's Department of Mechanical Engineering, and Seoul National University's robotics clubs SHAPE and SIGMA. Students who share a passion for robotics from the two universities joined forces across institutional lines, handling every stage themselves — from recruiting participants to designing the mission, running the event, and setting up the presentation and judging format. KAIST's Department of Mechanical Engineering supports the event with facilities and operational assistance so that the students' initiative can translate into genuine educational value.
Ten teams — 30 students total — selected from the two universities will take part. After receiving training in power circuits and robot joint control on August 3 and 4, participants will begin building their robots when the mission is unveiled on the morning of August 5 and continue working until 4 p.m. on August 8. Starting from an idea, they will go through design, assembly, programming, and repeated testing to complete a working robot — experiencing the full roughly 100-hour cycle themselves.
Each team will be provided with Angel Robotics' "phact" actuator, which serves as the robot's joints and muscles, and NVIDIA's Jetson AGX, which functions as the robot's brain. Taejin Technology Co., Ltd will provide training on circuits and electronic components for supplying stable power to the robots, and Angel Robotics will support hands-on training in using the actuators and controlling the robots.
Participants will not simply assemble a finished kit — they will design the robot's shape and movement from the ground up and build it themselves.
For fairness, the mission will be revealed only at the start of the hackathon on August 5. There is no single correct answer or predetermined robot form. Each team will interpret the same mission differently, combining mechanical structure, circuitry, AI, and control software into a single robot. One of the highlights will be seeing the different solutions the ten teams develop in response to the same mission.
The final day, August 8, will be an open Physical AI festival that welcomes the general public. A public conference at the KI Building (E4) Fusion Hall will introduce the current state of Physical AI in an accessible and engaging way — from robotic skin that lets robots feel touch like humans, to humanoid robots that can see and hear people, to a quadrupedal robot that has completed a marathon.
Professors Jung Kim, Yong-Hwa Park, and Jemin Hwangbo of the Department of Mechanical Engineering, along with Joon-Ha Kim, CEO of Diden Robotics, will each give a talk on robotic skin and haptics, multimodal perception in humanoid robots, the quadrupedal robot Raibo, and the journey of developing Physical AI for industrial use, respectively.
After the conference, an open demo day for the ten participating teams will run from 4 to 5 p.m. at KAIST's Culture Complex (E9), 3rd floor. Members of the public will be able to visit each team's booth, watch the robots the students built over 100 hours in action, and submit their own evaluations via QR code.
Judging criteria include mission achievement, technical execution, creativity, and presentation and demonstration. The final score will weight faculty advisor evaluation at 30%, peer evaluation among hackathon participants at 30%, sponsor judging panel evaluation at 30%, and pre-registered public attendee evaluation at 10%.
Angel Robotics and Taejin Technology are taking part as core technology partners, providing equipment and training. Faculty members and industry experts are providing education and technical guidance so that students can safely handle equipment used in real research and industrial settings.
"Physical AI's competitiveness comes not just from software but from hardware and control technology," said Kyoungchul Kong, Professor from Mechanical Engineering at KAIST and Head of the Future Technology Institute at Angel Robotics. "This hackathon will show that with high-performance robot components and the right development environment in place, even undergraduates can turn their imagination into a working robot in 100 hours," he added.
"This hackathon is about learning and building together, rather than competing between schools," said Yeonsu An, President of RoboticUS (and President of MR, KAIST's robotics club). "We hope the general public will get to experience the robots students have built firsthand and take part in the judging, coming away with the sense that Physical AI is a technology anyone can understand and enjoy — not just something for experts," she added.
“Although we do not yet know what the challenge will be, I am most looking forward to gathering in one place and developing robots together,” said Hyeontae Jeon, a participating student from Seoul National University. “It will be even more meaningful to work through challenges across university boundaries and present the robots we built ourselves to the public.”
The public conference is open to anyone through pre-registration. Pre-registered attendees can watch the lectures, view the open demo day, and take part in on-site judging. Registration is available on the RoboticUS official website or on Event-us, under "First Robot Hackathon – Robot & AI Public Conference (KAIST × SNU)." Registration closes August 6.
2026.08.04 View 607 -
KAIST Develops AI That Generates Feasible Plans for Delivery, Production, and Workforce Scheduling
From parcel delivery routes and factory production schedules to hospital duty rosters, many real-world planning tasks require solutions that satisfy numerous operational constraints. KAIST researchers have developed an artificial intelligence technique that can independently generate feasible plans satisfying all constraints specified in a mathematical optimization problem.
KAIST (President Choongsik Bae) announced on August 3 that a research team led by Professor Min-Soo Kim from the School of Computing has developed RL-SPH (Reinforcement Learning-based Start Primal Heuristic), a reinforcement learning technique that trains AI to independently produce feasible plans without relying on an external solver.
The key feature of the technology is its ability to learn how to produce solutions that satisfy the multiple constraints encoded in an optimization problem. The research team expects the method to serve as an important foundation for AI-based decision-making in fields including logistics, manufacturing, semiconductor production, and workforce management.
Parcel delivery routing, vehicle routing, factory production scheduling, and hospital staff rostering are representative planning problems that can be formulated using integer linear programming, or ILP. ILP is a mathematical optimization technique for finding the most efficient solution while satisfying a set of linear constraints and requiring some or all decision variables to take integer values.
A parcel delivery plan, for example, must do more than simply minimize delivery time. It must also comply with vehicle capacity limits and driver working-hour requirements while ensuring that every destination is visited. A route that violates even one of these conditions cannot be used in practice, regardless of how short or inexpensive it may appear.
Existing learning-based approaches can rapidly generate approximate or partial solutions, but these predictions frequently violate constraints. Consequently, many approaches pass their outputs to specialized ILP solvers, such as Gurobi or SCIP, which are then responsible for obtaining a feasible solution. The paper notes that existing end-to-end learning-based primal heuristics generally struggle to generate feasible solutions independently.
RL-SPH addresses this limitation by iteratively revising a candidate solution rather than attempting to predict the final answer in a single step. At each stage, it selects multiple decision variables that are likely to improve feasibility and determines whether their values should be increased, decreased, or left unchanged. The model then learns from the resulting changes in constraint violations and solution quality.
Notably, the team designed the AI to first find a plan that is actually usable, rather than the single best plan. The overall procedure consists of two stages. In the first stage, the AI prioritizes finding an initial feasible solution that satisfies all constraints. In the second stage, it seeks a higher-quality solution by reducing the objective value, such as cost or processing time, while maintaining feasibility.
For example, in a factory production-planning problem, the method would first identify a schedule that satisfies requirements such as delivery deadlines, equipment capacity, and available labor. It would then attempt to reduce production cost and time without violating those conditions. The research therefore prioritizes finding a plan that can actually be implemented before attempting to optimize it further.
The team also introduced ILP-GT, a new AI model that learns the relationships between variables and constraints, along with a feasibility-aware search strategy that prioritizes revising the variables most effective for resolving the problem, substantially improving computational efficiency.
Across five representative benchmarks, RL-SPH achieved a 100% feasibility rate, successfully finding a usable plan for every problem. It maintained the same performance even on more complex problems involving general (non-binary) integer variables.
Compared with existing techniques, RL-SPH reduced the primal gap — the gap between a method's solution and the best-known solution — by an average of 28.6 times, and improved the primal integral — a measure of the speed and quality of the search process — by 2.6 times. The time needed to find the first feasible plan was also 2.5 times faster on average.
Among recent AI techniques such as PAS, DDIM, and DiffILO, RL-SPH was the only method to achieve a 100% feasibility rate across three benchmarks compared (SC, CA, IS). Its training also took an average of just 30 minutes — 14.7 times faster than existing techniques and roughly 34 times faster than the most recent unsupervised learning — an AI training method that finds patterns in data without being given the correct answers in advance — based technique.
The technique further demonstrated its generalization potential on MIPLIB, an international benchmark library for mixed-integer programming widely used in academia and industry. It reliably found feasible plans not only for problems up to 67 times larger than those it was trained on, but also for entirely new problem types it had never encountered during training.
“In real-world applications, a plan that can actually be implemented is often more important than a theoretically optimal answer that violates practical constraints,” said Professor Kim.
He added, “This research demonstrates that AI can learn to generate feasible solutions without relying on a specialized optimization solver to enforce feasibility. We expect the technology to provide an important foundation for AI-based decision-making in logistics, manufacturing, semiconductor production, workforce management, and other industrial fields.”
Tae-Hoon Lee, a doctoral student in the KAIST School of Computing, participated as the first author, and Professor Min-Soo Kim led the research.
The findings were presented at the 43rd International Conference on Machine Learning, or ICML 2026, held in Seoul from July 6 to 11. ICML is regarded as one of the world’s premier international conferences in machine learning.
Paper title: RL-SPH: Learning to Achieve Feasible Solutions for Integer Linear Programs
DOI: https://doi.org/10.48550/arXiv.2411.19517
Authors: Tae-Hoon Lee (KAIST, first author), Min-Soo Kim (KAIST, corresponding author)
This research was supported by the Ministry of Science and ICT and the Institute of Information & Communications Technology Planning & Evaluation through related software research and Information Technology Research Center programs, as well as by the National Research Foundation of Korea. The paper’s acknowledgements specifically identify NRF and IITP support, including an ITRC grant.
2026.08.03 View 558 -
KAIST Develops AI That Avoids Hallucinating Even at Night or in Smoke
Multimodal large language models (MLLMs), which process multiple types of sensory information such as text, images, and audio at the same time, are rapidly expanding the range of applications for artificial intelligence (AI). However, in real-world environments, these models can misinterpret the physical characteristics of sensors, mistakenly identify objects, or claim to hear sounds that are not actually present simply because a certain object appears in a video. These errors are known as hallucinations. A KAIST research team has developed a new technology that corrects such information confusion and physical misperceptions in AI.
KAIST (President Choongsik Bae) announced on the 31st of July that a research team led by Professor Yong Man Ro from the School of Electrical Engineering has developed two core technologies that overcome the tendency of existing large language models to rely too heavily on ordinary camera (RGB) images and enable AI to suppress cross-modal hallucinations that occur when different sensory inputs become mixed.
The first technology developed by the research team is the Diverse Negative Attributes (DNA) optimization method, which helps AI accurately understand the physical characteristics of special camera sensors such as thermal, depth, and X-ray sensors. Existing AI models often failed to understand the physical meaning of such images, for example by mistaking bright areas in thermal images for simple light reflection.
The research team built VS-TDX, the first comprehensive benchmark for evaluating diverse vision sensors, and used the types of wrong answers that AI frequently produces as learning signals to help the model internalize the characteristics of each sensor. As a result, the AI gained a “new eye” that allows it to accurately infer the state of objects even in darkness or smoke.
The second technology is Modality-Adaptive Decoding (MAD), a control method that blocks hallucinations caused by confusion between visual and auditory information at the source. This technology prevents AI from mistakenly claiming that it hears a sound that does not actually exist simply because a certain object appears in a video.
MAD works by having the AI self-assess whether vision or audio is more important for a given task, and then increasing the weight of the more relevant modality in real time. A key advantage of this technology is that it can immediately suppress hallucination errors without costly model retraining, as it is training-free.
Instead of retraining AI models at large scale with massive computing resources, the research team maximized cost efficiency by introducing the DNA method, which enables fine adjustment with only a small amount of data, and the
MAD plug-in approach, which requires no additional training at all.
These technologies can be applied to autonomous vehicles operating at night or in bad weather, robots performing missions in smoke-filled environments, and unmanned aerial vehicles using thermal cameras. They are also expected to be useful in fields that process multiple types of sensor information together, such as airport X-ray security screening and medical image analysis.
Professor Yong Man Ro said, “This research is significant because it reduces AI’s sensory bias and misperceptions without large-scale retraining,” adding, “It will serve as a foundation for building multimodal AI that can be trusted in real-life and industrial settings.”
This achievement was notable for its continuity, with Sangyun Chung, a doctoral student in KAIST’s School of Electrical Engineering, participating as first author in both studies. Dr. Youngjun Yoo also participated as co-first author in the DNA study.
Among the related papers, the MAD study was presented in June at the Conference on Computer Vision and Pattern Recognition (CVPR), the world’s leading international conference in AI and computer vision. The DNA study was published in IEEE Transactions on Image Processing, a leading international journal in the field of image processing.
Paper title: Enhanced Vision-Language Models for Diverse Sensor Understanding: Cost-Efficient Optimization and Benchmarking,
DOI: 10.48550/arXiv.2412.20750 Author information: Sangyun Chung (KAIST, co-first author), Youngjun Yoo (KAIST, co-first author), Se Yeon Kim (KAIST, third author), Youngchae Chee (KAIST, fourth author), Yong Man Ro (KAIST, corresponding author)
Paper title: MAD: Modality-Adaptive Decoding for Mitigating Cross-Modal Hallucinations in Multimodal Large Language Models,
DOI: 10.48550/arXiv.2601.21181
Author information: Sangyun Chung (KAIST, first author), Se Yeon Kim (KAIST, second author), Youngchae Chee (KAIST, third author), Yong Man Ro (KAIST, corresponding author)
Related demo video: https://youtu.be/VuP9i6Vfk8o
This research was supported by the Institute of Information & Communications Technology Planning & Evaluation’s (IITP’s) Human-Centered AI Core Technology Development Program and by a Center for Applied Research in Artificial Intelligence (CARAI) grant funded by the Defense Acquisition Program Administration (DAPA) and the Agency for Defense Development (ADD).
2026.08.03 View 513 -
KAIST Develops AI That Finds Its Own Hidden Weaknesses, Paving the Way for Safer Generative AI Models
KAIST researchers have developed a safety verification technology that uncovers roughly seven times more hidden vulnerabilities in AI than existing methods. The technology is expected to serve as a foundation for developing safer, more trustworthy AI.
KAIST (President Choongsik Bae) announced on the 30th of July that a research team led by Professor Junmo Kim from the School of Electrical Engineering has developed a new framework called Stable-GFlowNet (S-GFN), which overcomes the limitations of red-teaming—a safety verification process that deliberately attacks large language models (LLMs) to expose hidden weaknesses.
Red-teaming for generative AI is the process of crafting attack prompts designed to probe an AI's vulnerabilities and induce the AI to produce harmful or dangerous responses before the program is deployed. Since discovering a wider variety of attack methods allows more vulnerabilities to be addressed in advance, both the success rate and diversity of attacks are critical.
Previous approaches primarily relied on reinforcement learning—an AI technique trained to maximize reward—to generate attack prompts. However, these methods frequently suffered from mode collapse—a phenomenon where the model repeatedly converges on a narrow set of high-reward attack prompts rather than generating diverse outputs, thereby limiting its ability to uncover various vulnerabilities.
Generative Flow Networks (GFlowNets)—an AI generation technique trained to produce diverse outputs in proportion to their reward—were proposed as a solution. Yet GFlowNet training is computationally complex and unstable, and noisy reward signals can assign high rewards even to meaningless sentences, often causing training to collapse.
To address these issues, the research team developed three core techniques that help the model learn effective attacks more reliably while filtering out flawed ones.
First, much like comparing several paths to choose the best one, the team introduced Contrastive Trajectory Balance (CTB), which reduces computational complexity and stabilizes training by directly comparing pairs of generated attack trajectories.
Second, akin to filtering out background noise to focus on a single voice, the team applied Noise Gradient Pruning (NGP) to eliminate minor reward fluctuations and ensure the model learns exclusively from meaningful signals.
Third, the team applied the Min-K Fluency Stabilizer (MKS), which guides the model to generate attack prompts resembling text that a real user would write—just as a human reader naturally prefers coherent sentences to gibberish.
As a result, Stable-GFlowNet discovered 134 unique attack types—about seven times more than the 17 unique attack types found by the existing GFlowNet-based technique—while maintaining a high attack success rate of 92%.
Defense models trained using attacks generated by Stable-GFlowNet also demonstrated strong generalization, effectively defending against a wide range of attacks in cross-attack tests, which evaluate performance using attack techniques different from those used during training.
The team further demonstrated that CTB and NGP achieve faster and more stable performance than existing methods—not only in AI safety verification, but also in other distribution-matching tasks such as molecular generation for drug discovery.
Professor Kim said, "This technology is significant in that it can reliably uncover a wide range of AI vulnerabilities even in realistic conditions with limited data and high noise." He added, "Because it allows a broader range of risk factors to be identified and defended against before generative AI is deployed in real-world services, we expect it to become a core foundational technology for developing safer, more trustworthy AI."
The study was led by first author Minchan Kwon, a Ph.D. candidate from the School of Electrical Engineering, and was selected as a Spotlight paper—placing it in the top 2.2% of submissions—at the International Conference on Machine Learning (ICML) 2026, one of the world's most prestigious AI conferences.
※ Paper title: Stable-GFlowNet: Toward Diverse and Robust LLM Red-Teaming via Contrastive Trajectory Balance
arXiv: https://arxiv.org/abs/2605.00553
This research was supported by the Institute of Information & Communications Technology Planning & Evaluation’s (IITP) SW Star Lab program, funded by the Ministry of Science and ICT.
2026.07.30 View 807 -
KAIST Develops AI That Learns to Theorize the World from Observation, Inspired by How Children Learn
A KAIST research team has developed a next-generation world model, an internal model an AI builds to understand and predict the world, that learns executable theories from observation alone.
KAIST (President Choongsik Bae) announced on the 15th of July that a team led by Professor Sungjin Ahn from the School of Computing has proposed a new learning paradigm called Learning-to-Theorize (L2T), which trains AI to theorize how the world works using only observed information. The team also built the Neural Theorizer (NEO), a neural network-based model that implements this paradigm.
The research was selected for an oral presentation at the 43rd International Conference on Machine Learning (ICML 2026), held in Seoul from July 6 to 11, and was presented on July 9. This places it among the top 0.7 percent (168 papers) of the 23,918 total submissions. The paper was also selected for the Best Paper Award at the Compositional Learning Workshop.
World models are a foundational technology across robot control, autonomous driving, generative AI, and autonomous agents — AI systems capable of judging and acting on their own. Until now, world models have mainly focused on predicting what happens next. Even when a model predicts the next scene accurately, it doesn't necessarily understand why that change occurred — that is, the underlying principle governing the world.
The team found a solution in how humans learn. Long before children acquire language, they build their own internal theories of how the world works. Applying this view from developmental cognitive science to AI, the researchers built a system that understands the principles behind the world, rather than one that simply predicts the future.
The team's proposed L2T framework provides no predetermined answers or rules. Given only a "before" and "after" observation, the AI discovers on its own which rule produced the change. While conventional AI models focus on "guessing what comes next," this approach is built to understand "why the change happened."
To implement this, the team also developed Neural Theorizier, NEO. The model discovers reusable primitives hidden within observed transformations and composes them into executable programs. These learned primitives can then be systematically recombined to explain new situations.
For example, NEO independently learns primitives corresponding to basic operations such as rotation, movement to the left or downward, and coloring. Even when presented with a combination it has never encountered during training, such as “move down, then color, then rotate”, it can recombine the primitives it has already learned to explain and solve the new situation.
Conventional AI, by contrast, tends to memorize entangled patterns as a single unit, so its performance drops sharply when faced with an unfamiliar combination. Through a range of experiments, the team demonstrated that NEO outperforms existing approaches in compositional generalization, the ability to combine learned basic rules to solve problems never seen before.
"It points to a new direction beyond prediction-centric world models, what we call a 'World Theory Model.' We expect this to develop into a core technology across fields including intelligent robots, autonomous agents, and AI that supports scientific discovery." said Professor Sungjin Ahn.
Master's students Doojin Baek and Gyubin Lee from the School of Computing served as co-first authors on the study.
Paper title: Learning to Theorize the World from Observation.
DOI: https://doi.org/10.48550/arXiv.2605.03413
Authors: Doojin Baek*, Gyubin Lee*, Junyeob Baek, Hosung Lee, Sungjin Ahn (*Co-first authors)
The research was supported by the National Research Foundation of Korea (NRF).
2026.07.27 View 1056 -
KAIST and NVIDIA Launch Human Physical AI NVAITC
A new era of Physical AI is taking shape, enabling wearable robots and humanoids to understand and predict human movement and achieve more precise control. KAIST, which possesses world-class research capabilities in wearable robotics, and NVIDIA will collaborate to develop a Human Motion Foundation Model that enables AI to learn human movement and physical intelligence.
KAIST, led by President Choongsik Bae, announced on July 25 that it will establish a NVIDIA AI Technology Center (NVAITC) with NVIDIA to advance collaborative research in Physical AI.
As the Korean government advances Physical AI as a key national initiative for the country’s future, the collaboration aims to secure core technologies for next-generation Physical AI by combining KAIST’s human-centered robotics technologies and real-world human motion data with NVIDIA AI technologies and global research network.
The collaboration will be carried out through the establishment of the Human Physical AI NVAITC by the KAIST Department of Mechanical Engineering and NVIDIA. The Human Physical AI Research Center at the KAIST Department of Mechanical Engineering will serve as the core research hub for the NVAITC . Building on this foundation, the two organizations plan to progressively expand the scope of their collaboration across the full spectrum of Physical AI, including wearable robots, humanoids, digital twins, and manufacturing.
“Competitiveness in the era of Physical AI will depend not simply on AI itself, but on domain-specific technologies and data grounded in a deep understanding of humans and robots,” said KAIST President Choongsik Bae. “By combining KAIST’s accumulated expertise in human-centered research with NVIDIA’s world-leading AI infrastructure and physical AI technologies, we will realize Physical AI that better understands and supports people and develop KAIST into a global hub leading Physical AI research and industry beyond Korea.”
The Human Physical AI Research Center was established around the laboratories of Professor Kyoungchul Kong, a leading researcher in wearable robotics, and Professor Jung Kim, a leading researcher in biorobotics. Professors Kim and Kong serve as co-directors of the Center.
The Center conducts research to understand how humans move, exert force, and maintain balance in real-world environments and to reproduce these capabilities through AI and robotics. In particular, the large-scale human motion data and gait and movement control technologies accumulated through wearable robotics research are regarded as a critical foundation for developing human-centered Physical AI.
Co-director Professor Kyoungchul Kong is a world-renowned researcher in wearable robotics who has developed robotic technologies for gait assistance and rehabilitation. Through Angel Robotics, a company he founded, he has also led the commercialization of wearable robotics by translating research outcomes into real-world products and services. Through the NVAITC , Professor Kong will lead the development of the Human Motion Foundation Model based on the human motion data and robotic control technologies accumulated by his research team.
On NVIDIA’s side, Charles Cheung, Senior Manager at the NVIDIA AI Technology Center (NVAITC), will participate by providing expert technical consultation and developer support. The NVAITC will also operate research and educational programs using NVIDIA Omniverse and digital twin platforms.
“The KAIST Human Physical AI Research Center has world-class human motion data and research capabilities in wearable robotics,” said Charles Cheung. “This research, which seeks to reproduce human movement through AI, is expected to open new possibilities for Physical AI.”
To ensure the systematic operation of the collaborative research, the two organizations will establish a Steering Committee and review research goals and progress every six months. They also plan to hold an annual international symposium that will bring together researchers from Korea and abroad to share the latest research outcomes and industry trends in Physical AI.
A Student Ambassador Program will also be offered to KAIST students. Through the program, NVIDIA experts will provide lectures and regular office hours and carry out projects with participating students.
The first cohort is expected to consist of five to 10 students. Participants will receive training focused on NVIDIA Omniverse and digital twin technologies and will be awarded certificates upon completion of the program.
The primary objective of the first phase of the collaborative research is to develop a Human Motion Foundation Model.
The Human Motion Foundation Model is a generative AI-based model trained on large-scale human motion data to understand, predict, and generate a wide range of human movements. It is expected to serve as a core enabling technology that will allow wearable robots and humanoids to more accurately identify users’ intentions and movements and respond more naturally.
The technologies developed through the NVAITC are expected to be applied not only to wearable robots that support the rehabilitation and daily lives of people with gait impairments, but also to humanoids that work alongside humans, human movement assessment, and digital healthcare.
Ultimately, the researchers aim to explain from an AI perspective how humans plan movement and control their muscles and joints. Based on this understanding, they seek to create next-generation robotic systems that help people overcome gait impairments and expand human physical capabilities.
The collaboration is also significant because its impact is expected to extend beyond an individual research project and contribute to the broader Physical AI industrial ecosystem in Korea.
Co-directors Professors Jung Kim and Kyoungchul Kong are currently leading in a Deep Tech Scale-up Valley project in the field of Physical AI. By combining Angel Robotics’ experience in technology commercialization, KAIST’s capabilities in robotics, mechanical engineering, and AI, and NVIDIA’s AI technologies and global professional network, the collaboration is expected to support a broad range of activities spanning research and development, talent cultivation, startup support, and technology commercialization.
“This collaboration will provide an important opportunity to take AI research in the Department of Mechanical Engineering to the next level,” said Professor Hyung-Soon Park, Head of the KAIST Department of Mechanical Engineering. “Centered around the Human Physical AI Research Center, we will expand Physical AI research into nationally strategic industries, including robotics and manufacturing.”
2026.07.26 View 2170 -
KAIST-NVIDIA Establish Asia's First AI Joint Research Lab, Accelerating Korea's AI Innovation
KAIST and NVIDIA establish Asia's first AI joint research lab between NVIDIA and a university to advance next-generation agentic AI tailored to the Korean language and domestic industries.
KAIST (President Choongsik Bae) announced on July 24 that it will establish the NVIDIA-KAIST Joint AI Research Lab at the Kim Jaechul Graduate School of AI with the global AI computing giant NVIDIA. The two organizations will conduct joint research on next-generation core AI technologies.
"This collaboration marks the starting point of a strategic partnership between KAIST and NVIDIA that combines world-class AI research talent with cutting-edge AI infrastructure," said President Choongsik Bae. “We will build Korea's leading global AI research hub and lead the way in developing next-generation foundational AI technologies and cultivating world-class AI talent."
Through this partnership, KAIST and NVIDIA will build a joint research framework covering agentic AI models and systems specialized for the Korean language and Korean industries. The two organizations will build a long-term research collaboration framework to develop foundational AI technologies for Korea, cultivate global talent, and strengthen industrial competitiveness.
The NVIDIA-KAIST AI Joint Research Lab, to be established at the Kim Jaechul Graduate School of AI, will operate as a global research hub developing agentic AI models and AI agent systems tailored to Korea's language and industrial needs.
The two organizations plan to combine NVIDIA’s full-stack AI technologies and Nemotron open models, and the computing infrastructure of local NVIDIA Cloud Partners with the world-class scientific talent at KAIST.
The $300 million collaboration will proceed over an initial five-year period, including $50 million per year in compute contributions. Researchers participating in the joint lab will gain access to the latest NVIDIA AI computing infrastructure through local NVIDIA Cloud Partners.
The joint lab will fund at least 10 KAIST researchers annually and provide each with internship opportunities at NVIDIA. In addition, NVIDIA plans to hire exceptional Korean researchers for full-time positions. Together, these efforts will create stronger pathways for Korea's top AI talent to pursue ambitious research, build long-term careers, and deepen global collaboration between academia and industry.
The joint lab will be led by Dr. Hyunwoo Kim, currently at NVIDIA, who will join the Kim Jaechul Graduate School of AI as a professor next month. Dr. Kim will set the lab's research direction and oversee collaboration between local and international researchers.
"Korea is home to leading AI researchers and has one of the world's most advanced technology ecosystems," said Bill Dally, chief scientist and senior vice president of research at NVIDIA. "The NVIDIA-KAIST AI Joint Research Lab will provide a foundation for pioneering the next frontier of AI research and accelerating the development of AI models and agent systems for Korea’s industries, language, and future"
Dr. Hyunwoo Kim, incoming faculty member at the KAIST Kim Jaechul Graduate School of AI, who will serve as head of the joint NVIDIA-KAIST lab upon joining KAIST, said,
“AI research is entering a new era — one that requires frontier talent, large-scale infrastructure and deep collaboration across academia and industry.” He added, “Together, NVIDIA and KAIST Kim Jaechul Graduate School of AI will pursue ambitious work that helps Korea attract and retain top AI scientists while building lasting ties with NVIDIA's global research organization.”
Song Chong, Head of the KAIST Kim Jaechul Graduate School of AI, said, "This joint lab is a new industry-academia collaboration model that combines world-class research talent, AI infrastructure, and the research capabilities of a global company." He added, "We will develop core agentic AI technologies specialized for the Korean language and Korean industries and build an ecosystem where outstanding researchers can carry out world-class research from within Korea."
Check out NVIDIA's official blog post on this historic partnership here.
2026.07.24 View 3581 -
KAIST Demonstrates World-Leading AI Capabilities for Solving Physics Problems with ICML 2026 Challenge Win
A KAIST research team has won an international AI challenge held in conjunction with ICML 2026 for developing a system that can interpret images and text, apply physical laws, and explain its reasoning. The result demonstrates KAIST’s world-leading capabilities in AI for understanding the physical world, with potential applications in aircraft design, robotics, autonomous vehicles, and space systems.
KAIST (President Choongsik Bae) announced on July 23 that a team led by Professor Il-Chul Moon from the KAIST AX Department won the Visual Grounded Physics Problem Solving Challenge, organized by the AI4Math Workshop, an official workshop of the International Conference on Machine Learning (ICML) 2026.
The competition ran from May 1 to June 16 and featured 139 teams, including participants from ETH Zurich, Fudan University, and the Shanghai Innovation Institute. The KAIST team achieved the highest score in the final evaluation and received the top prize at the ICML award ceremony held in Seoul on July 11.
The challenge evaluated whether AI could understand physics problems presented through images and text, apply physical laws, and generate both correct answers and the reasoning process. Participating systems had to integrate multimodal information and use principles such as Newtonian mechanics to solve problems logically. The focus was not simple calculation, but the ability to understand and explain complex physical phenomena.
The achievement suggests that AI can move beyond solving test questions to understanding real physical environments and optimizing their design and operation. The technology could support aircraft, robotics, autonomous driving, and space systems, where decisions must account for real-world physical conditions.
Five researchers from KAIST’s Applied Artificial Intelligence Laboratory (AAILab) participated, including doctoral student Jiseok Kwak from the Department of Industrial and Systems Engineering. The team built a multi-agent AI architecture in which several foundation models operated as independent agents and verified and debated one another’s answers.
This approach reduced errors from individual models and improved reasoning reliability, enabling the team to achieve the competition’s best performance. It also demonstrated new possibilities for AI reasoning concerning complex physical phenomena and agentic AI systems.
“By transforming foundation models into agents and observing multiple agents reach correct answers through debate and verification, we realized that no single foundation model will monopolize the future,” said Professor Moon.
“Although this competition took the form of a physics problem-solving test, the underlying technology is about developing AI for the optimal design and operation of physical systems, such as AI-based aircraft design and operations being explored by Boeing,” he added.
The research was supported by the Institute of Information & Communications Technology Planning & Evaluation and the ITRC Defense Swarm Systems Research Center.
2026.07.23 View 998