%73%63%68%6f%6f%6c%20%6f%66%20%63%6f%6d%70%75%74%69%6e%67
-
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 347 -
KAIST Professor Sooel Son Selected for Microsoft Funding for AI Safety and Security Research
KAIST (President Choongsik Bae) announced on the 28th of July that Professor Sooel Son has been selected as the only researcher in Korea to receive funding from Microsoft for research on artificial intelligence safety and security.
The funding was awarded through the External Red Team Alliance (EXTRA), a new program established by Microsoft’s AI Red Team, which examines the safety and security vulnerabilities of AI systems. EXTRA is a global initiative designed to strengthen AI safety and security research capabilities by supporting researchers at universities and technology experts around the world.
Microsoft noted that most AI safety testing is still conducted internally by individual companies or organizations. However, assessing the major risks posed by increasingly advanced AI systems requires expertise across a broad range of fields, including cybersecurity, multilingual environments, regional and cultural contexts, AI alignment, and potential misuse. EXTRA was launched in recognition of the difficulty a single internal organization faces when comprehensively evaluating these diverse risks.
Through the program, Microsoft will provide KAIST with an unrestricted gift of USD 25,000, approximately KRW 37 million, with no prescribed project period, to support research related to AI safety, security, alignment, and responsible AI development. The funding will be used to support Professor Son’s research team in its work on AI security and safety.
More than a dozen universities across six continents are participating in EXTRA, with KAIST being the only university selected from Korea. Through the program, Microsoft plans to expand the ecosystem for independent AI safety research and strengthen collaboration between academia and industry.
Professor Son’s research team has been conducting research on the security and privacy of AI systems that use machine-learning models and large language models. In particular, the team analyzes adversarial attacks against deep neural networks and language models—including model extraction, membership inference, personal information extraction, model inversion, machine unlearning, and prompt injection—and develops defense methodologies to assess and improve model safety.
Building on these technologies, the team is also focusing on establishing systematic defense methodologies that enable the safe and trustworthy deployment of agentic AI systems operating in real-world service environments, including web agents and agentic browsers.
“As AI systems rapidly spread throughout society, research that verifies the security and reliability of increasingly advanced AI is becoming more important,” said Professor Son. “Our participation in Microsoft AI Red Team’s EXTRA program will provide an opportunity to further advance our research on the safe development and use of AI systems.”
“AI safety research has never been more important, and universities have a critical role to play in advancing the field,” said Ram Shankar Siva Kumar, who leads the Microsoft AI Red Team. “Through EXTRA, we aim to support researchers working to deepen our understanding of how increasingly powerful AI systems can be evaluated, protected, and governed responsibly.”
“Competition in AI technology is expanding beyond performance to encompass safety and trustworthiness,” said KAIST President Choongsik Bae. “KAIST’s participation as the only Korean research institution in Microsoft’s global AI safety research network is a meaningful achievement that demonstrates Korea’s competitiveness in AI research. We will continue to lead the advancement of responsible AI technologies that everyone can trust and use by pursuing world-class research in AI safety and security.”
2026.07.28 View 1011 -
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 891 -
KAIST Develops Robot Learning Technology Capable of Precisely Imitating Even “Rough” Demonstrations
Robots with increasingly precise dexterity are becoming essential in everyday life and industrial settings, from assembling tiny smartphone components to assisting doctors in surgery. However, teaching robots delicate human movements has traditionally required collecting vast amounts of data at extremely fine time intervals, resulting in significant costs and time burdens. KAIST researchers have developed a robot artificial intelligence technology that can perform sophisticated tasks by autonomously adjusting precision according to the situation, even when trained only on coarsely (sparsely) sampled demonstrations.
KAIST, led by President Kwang Hyung Lee, announced on the 24th that a research team led by Professor Daehyung Park of the School of Computing has developed DiSPo, a multi-granularity manipulation model that generates fine-grained robot motions tailored to a user’s desired level of precision, even from rough human demonstrations.
Existing robot learning methods, such as Behavior Transformer and Diffusion Policy, are limited by their dependence on the time intervals of the data used during training. As a result, learning precision manipulation tasks such as screw fastening or component insertion has required collecting large volumes of high-frequency data at very short time intervals. This has significantly increased data collection costs and slowed down the inference speed of robot AI models.
To overcome these limitations, the research team combined Mamba, a state-space model capable of predicting time intervals, with a diffusion model that enables rich action representation. The team also introduced a new Step-scale factor mechanism, which allows users to directly control the time intervals used by the robot.
As a result, even when trained on only low-frequency (coarse) demonstration data, the robot can generate high-precision motions during inference without additional training by autonomously subdividing actions through a discretization process.
DiSPo achieved up to an 81% higher task success rate compared to state-of-the-art models in simulation environments. In real-world experiments using a collaborative robot, DiSPo stably performed challenging tasks such as passing a clamp through a narrow gap with only a 2.5 mm radial clearance and accurately pressing a small shutter button on a smartphone. This performance was up to four times higher than that of existing AI models.
The technology is expected to make a significant contribution to automation in a wide range of everyday and industrial service fields that require high precision, including precision component assembly, cable connection, medical surgery, and precision machining.
“This study demonstrates that robots can learn precise motions from coarse demonstrations and autonomously adjust their level of precision according to the task situation,” said Professor Daehyung Park. “Moving forward, this technology is expected to dramatically reduce data collection costs while serving as a general-purpose robot learning technology for various industrial fields, including precision assembly and medical applications.”
The study was led by Nayoung Oh, a master’s student at the KAIST Graduate School of AI, as the first author, and was presented on June 1 at the 2026 IEEE International Conference on Robotics and Automation, or ICRA 2026, one of the world’s most prestigious robotics conferences, held in Vienna, Austria.
Paper Title: DiSPo: Diffusion-SSM based Policy Learning for Coarse-to-Fine Action Discretization
DOI: https://doi.org/10.48550/arXiv.2409.14719
2026.06.24 View 1286 -
KAIST Develops Next-Generation Database Technology That Reduces AI Hallucinations and Improves Accuracy by 78%
One of the greatest weaknesses of AI agents that read and understand vast amounts of enterprise data is “hallucination” — the generation of plausible-sounding but factually incorrect information. KAIST researchers have developed a next-generation database technology capable of understanding documents, data, and relationships among entities all at once. The technology improves AI response accuracy by up to 78% and processing speed by up to 20 times, addressing a key challenge in the commercialization of enterprise AI.
KAIST, led by President Kwang Hyung Lee, announced on the 19th that a research team led by Professor Min-Soo Kim of the School of Computing, in collaboration with faculty startup GraphAI Co., Ltd., has developed “AkasicDB,” a next-generation database technology that integrates the functions of vector databases, graph databases, and relational databases into a single database management system (DBMS). Based on this technology, the team also developed a new Retrieval-Augmented Generation (RAG) method called “Omni RAG.”
AkasicDB is designed to integrate and execute the functions of vector databases, which convert the meaning of documents or images into numerical vectors to search for similar information; graph databases, which store and analyze relationships among entities such as people, companies, and products; and relational databases, which systematically manage data in table form. Omni RAG, developed on this foundation, improves the accuracy of generative AI responses by simultaneously utilizing semantic information from documents, relationships among entities, and structured data.
AI agents have recently been spreading rapidly based on RAG technology, which searches vast collections of corporate documents and expert knowledge and generates responses based on the retrieved information. However, real-world enterprise data is distributed across various forms, including documents, tables, and relationships among entities, making it difficult for AI to comprehensively understand and use the data. As a result, AI may generate factually incorrect responses without sufficient grounding, creating hallucination issues that have been regarded as a major obstacle to the broader adoption of enterprise AI.
Conventional RAG typically works by converting user queries and documents into vectors, retrieving semantically similar documents, and providing them to a large language model (LLM), an AI model trained on massive datasets to generate human-like language. While this approach is effective for searching unstructured documents, it has limitations when handling complex queries that must also consider relationships among entities in documents or structured conditions such as specific periods, types, or ranges.
For example, a query such as, “Find clauses related to Company A among contracts signed last year, and explain how those clauses are connected to product supply issues,” requires vector search to identify document meaning, graph search to explore relationships among entities, and relational queries to filter by date and type. In existing systems, this required building multiple types of databases separately and combining the results at the application layer, leading to management complexity and response delays.
To solve this problem, the research team proposed Omni RAG, which integrates vector similarity search, graph traversal, and relational filtering within a single query and execution plan. Omni RAG identifies more accurate evidence by simultaneously utilizing semantic information from documents, relationship information from knowledge graphs, and structural conditions from tabular data, significantly reducing AI hallucinations.
AkasicDB, developed to support this method, adopts a new architecture that integrates graph databases, vector databases, and relational databases into a single engine. Users can express complex RAG queries that combine vector search, graph traversal, and relational filtering as a single SQL/GQL* query, and AkasicDB optimizes and processes the query as one unified execution plan.
SQL/GQL, or Structured Query Language/Graph Query Language, refers to command languages used to search or modify information stored in databases. SQL is the traditional language used to handle tabular data, while GQL is a language dedicated to graph data and is used to analyze connections among entities such as people, companies, and products.
Through this integrated architecture, AkasicDB minimizes unnecessary intermediate result generation and data movement, greatly reducing the number of tokens used by LLMs and shortening response latency. In experiments, complex search queries that took up to 21.3 seconds in existing systems were processed in under one second, achieving a performance improvement of more than 20 times. Omni RAG also improved response accuracy by up to 78% compared with conventional RAG. These results demonstrate its potential to substantially mitigate hallucination, one of the core challenges for enterprise AI agents.
Professor Min-Soo Kim said, “For AI agents to accurately understand and utilize the vast amounts of data held by enterprises, data infrastructure capable of processing vector, graph, and relational data in an integrated manner within a single system is essential. AkasicDB is a next-generation database technology for the era of AI agents, and we expect it to be used as core data infrastructure in fields requiring high reliability, including defense, manufacturing, finance, law, science, and technology.”
KAIST School of Computing Ph.D. student Geonho Lee participated in this research as the first author. The research results were presented as a demo paper on June 2 at ACM SIGMOD 2026, one of the world’s most prestigious international conferences in the field of databases, where they drew strong interest from global companies and researchers.
※ Paper title: AkasicDB: Demonstrating Omni RAG with a Unified Vector-Graph-Relational DBMS
DOI: https://doi.org/10.1145/3788853.3801609
※ Author information: Geonho Lee, KAIST, first author; Jeongho Park and Donghyoung Han, GraphAI Co., Ltd., co-authors; Professor Min-Soo Kim, KAIST, corresponding author
※ Demonstration video: https://www.youtube.com/watch?v=KD6MznZ61P4
2026.06.19 View 1929 -
Professor Insik Shin Becomes First Korean to Win the RTSS Most Influential Paper Award
< KAIST Professor Insik Shin >
KAIST announced on December 21st that Professor Insik Shin from the School of Computing has received the Influential Paper Award 2025 at the IEEE Real-Time Systems Symposium (RTSS), the world's most prestigious international conference in the field of real-time systems.
This honor is a "Test of Time Award," presented to papers that have exerted a sustained and significant influence on both academia and industry for more than 10 years after publication. This marks the first time a Korean researcher has received this prestigious award. The ceremony took place at IEEE RTSS 2025 in Boston, USA, on December 4th (local time).
Professor Shin’s award-winning research is the "Periodic Resource Model," co published in 2003 with Professor Insup Lee of the University of Pennsylvania. Rather than trying to verify a complex machine or system all at once, this study developed a method to verify individual components—much like LEGO blocks—to ensure each meets its designated timing requirements. It mathematically guarantees that when these components are assembled, the entire system will operate safely.
Paper Title: Periodic Resource Model for Compositional Real-Time Guarantees
DOI: 10.1109/REAL.2003.1253249
Thanks to this research, it has become possible to design real-time systems that cannot tolerate even a moment of delay—such as autonomous vehicles, aircraft, and industrial robots—with greater precision and safety. This breakthrough overcame the limitations of existing methods that required analyzing an entire system at once, which had become nearly impossible as the complexity of modern real-time systems increased rapidly.
Professor Shin presented a method to divide a system into small modules, verify that each module satisfies its time constraints, and mathematically prove that the safety of the entire system is guaranteed upon integration. This work is credited with establishing the foundation for modern compositional real-time scheduling theory.
At the time of its initial publication in 2003, this paper won the 'Best Paper Award' at RTSS—another first for a Korean researcher. Now, 20 years later, its academic and industrial value has been officially recognized once again. This is because the theory has transcended academic boundaries to become a core analytical tool in various safety-critical industries, including autonomous driving, aerospace control, and industrial automation.
The IEEE Technical Committee stated, "This model has established itself as a core language for modern real-time system design and has guided the direction of research and industry for the past 20 years." The paper is currently featured in textbooks at major universities in the United States and Europe, serving as a standard theory in the field.
"As a scholar, this is the award I have wanted most in my life," said Professor Shin. "I am honored to have it recognized that research from 20 years ago has truly had a major impact on the world. This was made possible by the many researchers and companies who applied this theory to actual systems."
Meanwhile, Professor Shin is expanding his research beyond real-time systems into the field of Artificial Intelligence (AI). He founded the faculty-led startup Fluiz and developed FluidGPT, a mobile AI agent technology that allows users to execute smartphone apps via voice commands. This technology recently won the AI Champion Competition hosted by the Ministry of Science and ICT. Experts evaluate Professor Shin as achieving rare success by bridging basic theory and applied technology, effectively linking research to industry.
2025.12.22 View 2477 -
AI Technology World No. 1 in Finding the Exact Moment in a Video: Where is the First Place?
< (From left) Professor Joon Hyuk Noh (Assistant Professor, Department of Artificial Intelligence, Ewha Womans University), Seojin Hwan, Yoonki Cho (Ph.D. Candidate), Professor Sung-Eui Yoon (School of Computing, KAIST) >
When faced with a complex question like 'What object disappeared while the camera was pointing elsewhere?', a common problem is that AI often relies on language patterns to guess a 'plausible answer,' instead of actually observing the real situation in the video. To overcome this limitation, our university's research team developed a technology that enables the AI to autonomously identify the 'exact critical moment (Trigger moment)' within the video, and the team’s excellence was proven by winning an international AI competition with this technology. The university announced on the 28th that the research team led by Professor Sung-Eui Yoon from the School of Computing, in collaboration with Professor Joon Hyuk Noh's team from Ewha Womans University, took 1st place in the Grounded Video Question Answering track of the Perception Test Challenge held at ICCV 2025, a world-renowned computer vision conference. The Perception Test Challenge held at ICCV 2025 was organized by Google DeepMind with a total prize pool of 50,000 Euros (approximately 83 million KRW). It assesses the cognitive and reasoning abilities of multimodal AI, which must comprehensively understand various data, including video, audio, and text. Crucially, the core evaluation factor is the ability to make judgments based on actual video evidence, moving beyond language-centric bias. Unlike conventional methods that analyze the entire video indiscriminately, our university's research team developed a new technology that instructs the AI to first locate the core scene (Trigger moment) essential for finding the correct answer. Simply put, this technology is designed to make the AI autonomously determine: “This scene is decisive for answering this question!” The research team calls this framework CORTEX (Chain-of-Reasoning for Trigger Moment Extraction). The research team's system consists of a three-stage structure where three models performing different functions operate sequentially. First, the Reasoning AI (Gemini 2.5 Pro) reasons about which moment is required to answer the question and finds candidate Trigger moments. Next, the Object Location Finding Model (Grounding Model, Molmo-7B) accurately identifies the exact location (coordinates) of people, cars, and objects on the screen during the selected moment. Finally, the Tracking Model (SAM2) precisely tracks the movement of objects in the time frame before and after the selected scene, using that scene as a reference, thereby reducing errors. In short, the 'method of accurately pinpointing a key scene and tracking the evidence for the answer centered on that scene' significantly reduced problems like initial misjudgment or occlusion in the video. In the Grounded Video Question Answering (Grounded VideoQA) track, which saw 23 participating teams, the KAIST team SGVR Lab (Scalable Graphics, Vision & Robotics Lab) recorded 0.4968 points in the HOTA (Higher Order Tracking Accuracy) metric, overwhelmingly surpassing the 2nd place score of 0.4304 from Columbia University, USA, to secure 1st place. This achievement is nearly double the previous year's winning score of 0.2704 points. This technology has wide-ranging applications in real-life settings. Autonomous driving vehicles can accurately identify moments of potential accident risk, robots can understand the surrounding environment smarter, security and surveillance systems can rapidly locate critical scenes, and media analysis can precisely track the actions of people or objects in chronological order. This is a core technology that enables AI to judge based on "actual evidence in the video." The ability to accurately pinpoint how objects behave over time in a video is expected to greatly expand the application of AI in real-world scenarios in the future.
< Pipeline image of the grounding framework for video question answering proposed by the research team >
This research was presented on October 19th at ICCV 2025, the 3rd Perception Test Challenge conference. The achievement was supported by the Ministry of Science and ICT's Basic Research Program (Mid-Career Researcher), the SW Star Lab Project's 'Development of Perception, Action, and Interaction Algorithms for Open-World Robot Services,' and the AGI Project's 'Reality Construction and Bi-directional Capability Approach based on Cognitive Agents for Embodied AGI' tasks."
2025.12.01 View 3224 -
Professor Youngjin Kwon's Team Wins Google Award 'Catches Bugs Without a Real CPU
< Professor Youngjin Kwon >
Modern CPUs have complex structures, and in the process of handling multiple tasks simultaneously, an order-scrambling error known as a 'concurrency bug' can occur. Although this can lead to security issues, these bugs were extremely difficult to detect using conventional methods. Our university's research team has developed a world-first-level technology to automatically detect these bugs by precisely reproducing the internal operation of the CPU in a virtual environment without needing a physical chip. Through this, they successfully found and fixed 11 new bugs in the latest Linux kernel.
Our university announced on the 21st that the research team led by Professor Youngjin Kwon of the School of Computing has won the 'Research Scholar Award' (Systems category) presented by Google.
The Google Research Scholar Award is a global research support program, implemented since 2020, to support Early-Career Professors conducting innovative research in various fields such as AI, Systems, Security, and Data Management.
It is known as a highly competitive program, with the selection process conducted directly by Google Research scientists, and only a tiny fraction of the hundreds of applicants worldwide are chosen. In particular, this award is recognized as one of the most prestigious industry research support programs globally in the field of AI and Computer Systems, and domestic recipients are rare.
■ Technology Developed to Detect Concurrency Bugs in the Latest Apple M3 and ARM Servers
Professor Kwon's team developed a technology that automatically detects concurrency bugs in the latest ARM (a CPU design method that uses less power and is highly efficient) based servers, such as the Apple M3 (Apple's latest-generation computer processor chip).
A concurrency bug is an error that occurs when the order of operations gets mixed up while the CPU handles multiple tasks simultaneously. This is a severe security vulnerability that can cause the computer to suddenly freeze or become a pathway for hackers to attack the system. However, these errors were extremely difficult to find with existing testing methods alone.
■ Automatically Detects Bugs by Reproducing CPU Internal Operations Without a Real CPU
The core achievement of Professor Kwon's team is the 'technology to reproduce the internal operation of the CPU exactly in a virtual environment without a physical chip.' Using this technology, it is possible to precisely analyze the order in which instructions are executed and where problems occur using only software, without having to disassemble the CPU or use the actual chip.
By running the Linux operating system based on this system to automatically detect bugs, the research team discovered 11 new bugs in the latest Linux kernel* and reported them to the developer community, where they were all fixed.
*Linux kernel: The core operating system engine that forms the basis of servers, supercomputers, and smartphones (Android) worldwide. It acts as the 'heart' of the system, managing the CPU, memory, and storage devices.
Google recognized this technology as 'very important for its own infrastructure' and conferred the Award.
< Google Scholar Award Recipient Page >
This technology is evaluated to have general applicability, not only to Linux but also to various operating systems such as Android and Windows. The research team has released the software as open-source (GitHub) so that anyone in academia or industry can utilize it.
Professor Youngjin Kwon stated, "This award validates the international competitiveness of KAIST's systems research," and "We will continue our research to establish a safe and highly reliable computing environment."
※ Google Scholar Award Recipient Page: https://research.google/programs-and-events/research-scholar-program/recipients/ GitHub (Technology Open-Source): https://github.com/casys-kaist/ozz
2025.11.21 View 3327 -
Automatic C to Rust Translation Technology Gains Global Attention for Accuracy Beyond AI
<(From Left) Professor Sukyoung Ryu, Researcher Jaemin Hong>
As the C language, which forms the basis of critical global software like operating systems, faces security limitations, KAIST's research team is pioneering core original technology research for the accurate automatic conversion to Rust to replace it. By proving the mathematical correctness of the conversion, a limitation of existing Artificial Intelligence (LLM) methods, and solving C language security issues through automatic conversion to Rust, they presented a new direction and vision for future software security research. This work has been selected as the cover story for CACM, the world's highest-authority academic journal, thereby demonstrating KAIST's global research leadership in the field of computer science.
KAIST announced on the 9th of November that the paper by Professor Sukyoung Ryu's research team (Programming Language Research Group) from the School of Computing was selected as the cover story for the November issue of CACM (Communications of the ACM), the highest authority academic journal published by ACM (Association for Computing Machinery), the world's largest computer society.
<Photo of the Paper Selected for the Cover of Communications of the ACM>
This paper comprehensively addresses the technology developed by Professor Sukyoung Ryu's research team for the automatic conversion of C language to Rust, and it received high acclaim from the international research community for presenting the technical vision and academic direction this research should pursue in the future.
The C language has been widely used in the industry since the 1970s, but its structural limitations have continuously caused severe bugs and security vulnerabilities. Rust, on the other hand, is a secure programming language developed since 2015, used in the development of operating systems and web browsers, and has the characteristic of being able to detect and prevent bugs before program execution.
The US White House recommended discontinuing the use of C language in a technology report released in February 2024, and the Defense Advanced Research Projects Agency (DARPA) also explicitly stated that Rust is the core alternative for resolving C language security issues by promoting a project to develop technology for the automatic conversion of C code to Rust.
Professor Sukyoung Ryu's research team proactively raised the issues of C language safety and the importance of automatic conversion even before these movements began in earnest, and they have continuously developed core related technologies.
In May 2023, the research team presented the Mutex conversion technology (necessary for program synchronization) at ICSE (International Conference on Software Eng), the top authority conference in software engineering. In June 2024, they presented the Output Parameter conversion technology (used for result delivery) at PLDI (Programming Language Design and Implementation), the top conference in programming languages, and in October of the same year, they presented the Union conversion technology (for storing diverse data together) at ASE (Automated Software Eng), the representative conference in software automation.
These three studies are all "world-first" achievements presented at top-tier international academic conferences, successfully implementing automatic conversion technology for each feature with high completeness.
Since 2023, the research team has consistently published papers in CACM every year, establishing themselves as global leading researchers who consistently solve important and challenging problems worldwide.
This paper was published in CACM (Communications of the ACM) on October 24, with Dr. Jaemin Hong (Postdoctoral Research Fellow at KAIST Information and Electronics Research Institute) as the first author. ※Paper Title: Automatically Translating C to Rust, DOI: https://doi.org/10.1145/3737696
Dr. Jaemin Hong stated, "The conversion technology we developed is an original technology based on programming language theory, and its biggest strength is that we can logically prove the 'correctness' of the conversion." He added, "While most research relies on Large Language Models (LLMs), our technology can mathematically guarantee the correctness of the conversion."
Dr. Hong is scheduled to be appointed as an Assistant Professor in the Computer Science Department at UNIST starting in March 2025.
Furthermore, Professor Ryu's research team has four papers accepted for presentation at ASE 2025, the highest-authority conference in software engineering, including C→Rust conversion technology.
These papers, in addition to automatic conversion technology, cover various cutting-edge software engineering fields and are receiving high international acclaim. They include: technology to verify whether quantum computer programs operate correctly, 'WEST' technology that automatically checks the correctness of WebAssembly programs (technology for fast and efficient program execution on the web) and creates tests for them, and technology that automatically simplifies complex WebAssembly code to quickly find errors. Among these, the WEST paper received the Distinguished Paper Award.
This research was supported by the Leading Research Center/Mid-career Researcher Support Program of the National Research Foundation of Korea, the Institute of Information & Communications Technology Planning & Evaluation (IITP), and Samsung Electronics.
2025.11.10 View 8151 -
3D Worlds from Just a Few Phone Photos
<(From Left) Ph.D candidate Jumin Lee, Ph.D candidate Woo Jae Kim, Ph.D candidate Youngju Na, Ph.D candidate Kyu Beom Han, Professor Sung-eui Yoon>
Existing 3D scene reconstructions require a cumbersome process of precisely measuring physical spaces with LiDAR or 3D scanners, or correcting thousands of photos along with camera pose information. The research team at KAIST has overcome these limitations and introduced a technology enabling the reconstruction of 3D —from tabletop objects to outdoor scenes—with just two to three ordinary photographs. The breakthrough suggests a new paradigm in which spaces captured by camera can be immediately transformed into virtual environments.
KAIST announced on November 6 that the research team led by Professor Sung-Eui Yoon from the School of Computing has developed a new technology called SHARE (Shape-Ray Estimation), which can reconstruct high-quality 3D scenes using only ordinary images, without precise camera pose information.
Existing 3D reconstruction technology has been limited by the requirement of precise camera position and orientation information at the time of shooting to reproduce 3D scenes from a small number of images. This has necessitated specialized equipment or complex calibration processes, making real-world applications difficult and slowing widespread adoption.
To solve these problems, the research team developed a technology that constructs accurate 3D models by simultaneously estimating the 3D scene and the camera orientation using just two to three standard photographs. The technology has been recognized for its high efficiency and versatility, enabling rapid and precise reconstruction in real-world environments without additional training or complex calibration processes.
While existing methods calculate 3D structures from known camera poses, SHARE autonomously extracts spatial information from images themselves and infers both camera pose and scene structure. This enables stable 3D reconstruction without shape distortion by aligning multiple images taken from different positions into a single unified space.
<Representative Image of SHARE Technology>
"The SHARE technology is a breakthrough that dramatically lowers the barrier to entry for 3D reconstruction,” said Professor Sung-Eui Yoon. “It will enable the creation of high-quality content in various industries such as construction, media, and gaming using only a smartphone camera. It also has diverse application possibilities, such as building low-cost simulation environments in the fields of robotics and autonomous driving."
<SHARE Technology, Precise Camera Information and 3D Scene Prediction Technology>
Ph.D. Candidate Youngju Na and M.S candidate Taeyeon Kim participated as co-first authors on the research. The results were presented on September 17th at the IEEE International Conference on Image Processing (ICIP 2025), where the paper received the Best Student Paper Award.
The award, given to only one paper among 643 accepted papers this year—a selection rate of 0.16 percent—once again underscores the excellent research capabilities of the KAIST research team.
Paper Title: Pose-free 3D Gaussian Splatting via Shape-Ray Estimation, DOI: https://arxiv.org/abs/2505.22978
Award Information: https://www.linkedin.com/posts/ieeeicip_congratulations-to-the-icip-2025-best-activity-7374146976449335297-6hXz
This achievement was carried out with support from the Ministry of Science and ICT's SW Star Lab Project under the task 'Development of Perception, Action, and Interaction Algorithms for Unspecified Environments for Open World Robot Services.'
2025.11.10 View 4090 -
KAIST's 'FluidGPT' Wins Grand Prize at the 2025 AI Champion Competition
<Commemorative Photo After Winning at the 2025 AI Champions Award Ceremony>
The era has begun where an AI assistant goes beyond simple conversation to directly view the screen, make decisions, and complete tasks such as hailing a taxi or booking an SRT ticket.
KAIST (President Kwang Hyung Lee) announced on the 6th that the AutoPhone Team (Fluidez, KAIST, Korea University, Sungkyunkwan University), led by Professor Insik Shin (CEO of Fluidez Co., Ltd.) of the School of Computing, was selected as the inaugural AI Champion (1st place) in the '2025 Artificial Intelligence Champion (AI Champion) Competition,' hosted by the Ministry of Science and ICT.
This competition is the nation's largest AI technology contest, comprehensively evaluating the innovativeness, social impact, and commercial potential of AI technology. With 630 teams participating nationwide, the AutoPhone Team claimed the top honor and will receive 3 billion Korean won in research and development funding.
The technology developed by the AutoPhone Team, 'FluidGPT,' is a fully autonomous AI agent that understands a user's voice command and enables the smartphone to independently run apps, click, input, and even complete payments.
For example, when a user says, "Book an SRT ticket from Seoul Station to Busan," or "Call a taxi," FluidGPT opens the actual app and sequentially performs the necessary steps to complete the request.
The core of this technology is its 'Non-Invasive (API-Free)' structure. Previously, calling a taxi using an app required directly connecting to the app's internal system (API communication) through the taxi app's API. In contrast, this technology does not modify the existing app's code or link an API. Instead, the AI directly recognizes and operates the screen (UI), acquiring the ability to use the smartphone just like a human.
As a result, FluidGPT presents a new paradigm—"AI that sees, judges, and moves a hand on behalf of a person"—and is evaluated as a core technology that will usher in the 'AI Phone Era.'
FluidGPT moves beyond simple voice assistance to implement the concept of 'Agentic AI' (Action-Oriented Artificial Intelligence), where the AI directly views the screen, makes decisions, and takes action. As a fully action-oriented system, the AI clicks app buttons, fills in input fields, and references data to autonomously achieve the user's objective, foreshadowing an innovation in how smartphones are used.
Professor In-sik Shin of the School of Computing shared his thoughts, stating, "AI is now evolving from conversation to action. FluidGPT is a technology that understands the user's words and autonomously executes actual apps, and it will be the starting point of the 'AI Phone Era.' The AutoPhone Team possesses world-class research capabilities, and we will contribute to the widespread adoption of AI services that everyone can easily use."
KAIST President Kwang Hyung Lee remarked, "This achievement is a representative example that demonstrates KAIST's vision for AI convergence," adding, "AI technology is entering the daily lives of citizens and leading a new wave of innovation." He further added, "KAIST will continue to lead research in future core technologies such as AI and semiconductors to bolster national competitiveness."
2025.11.06 View 4460 -
KAIST, Dancing Like 'Navillera'... AI Understands and Renders Garment Motions of Avatars
<(From Left)Ph.D candidate Jihyun Lee, Professor Tae-Kyun Kim, M.S candidate Changmin Lee>
The era has begun where AI moves beyond merely 'plausibly drawing' to understanding even why clothes flutter and wrinkles form. A KAIST research team has developed a new generative AI that learns movement and interaction in 3D space following physical laws. This technology, which overcomes the limitations of existing 2D-based video AI, is expected to enhance the realism of avatars in films, the metaverse, and games, and significantly reduce the need for motion capture or manual 3D graphics work.
KAIST (President Kwang Hyung Lee) announced on the 22nd that the research team of Professor Tae-Kyun (T-K) Kim from the School of Computing has developed 'MPMAvatar,' a spatial and physics-based generative AI model that overcomes the limitations of existing 2D pixel-based video generation technology.
To solve the problems of conventional 2D technology, the research team proposed a new method that reconstructs multi-view images into 3D space using Gaussian Splatting and combines it with the Material Point Method (MPM), a physics simulation technique.
In other words, the AI was trained to learn physical laws on its own by stereoscopically reconstructing videos taken from multiple viewpoints and allowing objects within that space to move and interact as if they were in real physical world.
This enables the AI to compute the movement based on objects' material, shape, and external forces, and then learn the physical laws by comparing the results with actual videos.
The research team represented the 3D space using point-units, and by applying both Gaussian and MPM to each point, they simultaneously achieved physically natural movement and realistic video rendering.
That is, they divided the 3D space into numerous small points, making each point move and deform like a real object, thereby realizing natural video that is nearly indistinguishable from reality.
In particular, to precisely express the interaction of thin and complex objects like clothing, they calculated both the object's surface (mesh) and its particle-unit structure (point), and utilized the Material Point Method (MPM), which calculates the object's movement and deformation in 3D space according to physical laws.
Furthermore, they developed a new collision handling technology to realistically reproduce scenes where clothes or objects move and collide with each other in multiple spots and complex manner.
The generative AI model MPMAvatar, to which this technology is applied, successfully reproduced the realistic movement and interaction of a person wearing loose clothing, and also succeeded in 'Zero-shot' generation, where the AI processes data it has never seen during the learning process by inferring on its own.
<Figure 1. Modeling new human poses and clothing dynamics from multi-view video input, and zero-shot generation of novel physical interactions.>
The proposed method is applicable to various physical properties, such as rigid bodies, deformable objects, and fluids, allowing it to be used not only for avatars but also for the generation of general complex scenes.
<“Figure 2. Depiction of graceful dance movements and soft clothing folds, like Navillera.>
Professor Tae-Kyun (T-K) Kim explained, "This technology goes beyond AI simply drawing a picture; it makes the AI understand 'why' the world in front of it looks the way it does. This research demonstrates the potential of 'Physical AI' that understands and predicts physical laws, marking an important turning point toward AGI (Artificial General Intelligence)." He added, "It is expected to be practically applied across the broaden immersive content industry, including virtual production, films, short-form contents, and adverts, creating significant change."
The research team is currently expanding this technology to develop a model that can generate physically consistent 3D videos simply from a user's text input.
This research involved Changmin Lee, a Master's student at the KAIST Graduate School of AI, as the first author, and Jihyun Lee, a Ph.D. student at the KAIST School of Computing, as a co-author. The research results will be presented at NeurIPS, the most prestigious international academic conference in the field of AI, on December 2nd, and the program code is to be fully released.
· Paper: C. Lee, J. Lee, T-K. Kim, MPMAvatar: Learning 3D Gaussian Avatars with Accurate and Robust Physics-Based Dynamics, Proc. of Thirty-Ninth Annual Conf. on Neural Information Processing Systems (NeurIPS), San Diego, US, 2025
· arXiv version: https://arxiv.org/abs/2510.01619
· Related Project Site: https://kaistchangmin.github.io/MPMAvatar/
· Related video links showing the 'Navillera'-like dancing drawn by AI:
o https://www.youtube.com/shorts/ZE2KoRvUF5c
o https://youtu.be/ytrKDNqACqM
This work was supported by the Institute of Information & Communications Technology Planning & Evaluation(IITP) grant funded by the Korea government(MSIT) through the Human-Oriented Next-Generation Challenging AGI Technology Project (RS-2025-25443318) and the Professional AI Talent Development Program for Multimodal AI Agents (RS-2025-25441313).
2025.10.22 View 5168