My paper titled “Beyond Hamming: Query-Aware Decoding of Binary Cosine Sketches” is accepted to ICML 2026. The paper is about angular distance estimator for vector DB like RAG. This study identifies the near-background region as the key bottleneck for retrieval performance in 1-bit binary-sketch-based cosine estimation, and proposes QA-Cos, a new decoder that improves estimation […]
We are thrilled to share the wonderful news that Professor Rhongho Jang, who completed his Ph.D. in our lab in March 2020, has received the prestigious NSF CAREER Award from the U.S. National Science Foundation (NSF). His awarded project is titled “CAREER: In-network GPU Integration for Vision-inspired Generative Inference.” The NSF CAREER Award is widely […]
Our group’s paper on defeating slow-and-low attacks was accepted in NSDI 2026! The title is “Defeating Slow-and-Low Threats via Diffusion Model-based Generative Inference”. Congratulations, Rhongho, Mehdi, and all.
Our education and research group, “Trustworthy and Resource-efficient Unified Evolving AI (TRUE-AI),” was awarded the BK21 FOUR program. The program provides us with the total budget of 1.2B KRW through February 2027, and we are preparing to transition to the BK21 FIVE program thereafter. The TRUE-AI group consists of 12 faculty members from Cybersecurity, Computer […]
Professor Nyang gave an interview to the Ewha Womans University Newspaper, where he shared his vision and aspirations as the Director of the Office of Information and Communications. For more details, check this out!
Owing to Mohammed and David‘s great efforts, our paper titled “SHIELD: Thwarting Code Authorship Attribution” is finally accepted in IEEE Transactions on Dependable and Secure Computing. This work lies in the line of works on source code authorship starting from CCS paper in 2018, but the aim of the work is quite opposite to that. […]
Arriving San Diego. A fine view from the hotel room. Omar, the session chair and Sian Sian giving a talk in the session Sian giving a talk in the session Our group just after Sian’s talk photographed by Prof. Heekuck Oh Me and Aziz by the pool Me […]
SketchFeature is a real-time high-order feature extractor for security in data plane of high speed routers. This work is the first 3rd-order feature extractor that is feasible in data plane. Congratulations to Sian, Mehdi, Bara, Rhongho!
A paper titled “Enhancing Network Attack Detection with Distributed and In-Network Data Collection System” is accepted at USENIX Security 2024. This paper explores a new possibility of distributed feature measurement framework in the era of AI using data plane programmability. The idea is to prioritize large flows over super mice to measure immediately in the […]
Now, we are BRL (Basic Research Lab) together with Prof. Dongbo Min and Prof. Se Eun Oh. Title: Research on SPV(Security/Privacy/Visual) Firewall for Connected Autonomous Vehicle Period: 2023.6-2026.2 Budget: 1.375 billion KRW Aim: In recent years, the amount of data exchanged between automobiles and external networks has increased exponentially due to big data, deep learning, […]
