Small representations.
Large-scale possibilities.

2003Lab established
156Publications
05Research areas

WHAT WE STUDY

Efficient systems. Secure foundations.

01

Approximate computing & data sketches

Compact data sketches for counting and high-order frequency estimation.

02

(Encrypted) traffic analysis

Measuring and analyzing encrypted traffic with efficient algorithms and programmable data planes.

03

Efficient KV Cache System

Angular distance and inner-product estimation for efficient KV caching, from our ICML work to LAS (Learned Attention System).

04

High-performance Key-Value Store

Exploring LSM-tree designs for high-performance key-value storage.

05

Differential privacy for AI

Privacy-preserving data sketches and resource-efficient AI systems.

The Information Security Research Laboratory (ISRL) studies how to measure and protect networks at scale, from in-network security on programmable routers to privacy-preserving machine learning. Explore our publications or the latest news.

OUR PEOPLE

Students

Yeawon You

PhD student

Jiyeon Kim

PhD student

Gayoung Kim

MS student

Hanui Choi

MS student

LATEST WORK

Recent publications

All publications
Conference2026

Beyond Hamming: Query-Aware Decoding of Binary Cosine Sketches

DaeHun Nyang

International Conference on Machine Learning (ICML 2026)

Conference2026

Defeating Slow-and-Low Threats via Diffusion Model-based Generative Inference

Seyed Mohammad Mehdi Mirnajafizadeh, Prashant Khanduri, DaeHun Nyang, Rhongho Jang

USENIX NSDI (USENIX Symposium on Networked Systems Design and Implementation) 2026

Journal2025

Enhancing Vulnerability Reports with Automated and Augmented Description Summarization

Hattan Althebeiti, Mohammed Alkinoon, Manar Mohaisen, Saeed Salem, DaeHun Nyang, David Mohaisen

IEEE Transactions on Big Data

BUILD WHAT COMES NEXT

Curious minds.
Meaningful research.

We welcome undergraduate interns and graduate students interested in AI, security, and scalable systems.

Join our lab