POSTECH · EE Department

Computing for Embedded Lightweight Learning — designing efficient, brain-inspired AI systems for the next generation of computing.

🔬 Explore Projects 🌐 Lab Website 📄 All Publications
Research Focus

Three Pillars of CELL

Our work sits at the intersection of efficient AI, brain-inspired computing, and system-level optimization — pushing the frontier of on-device intelligence.

🧠
On-Device Generative AI
Large Language Models, Diffusion models, and efficient LLM adaptation techniques including pruning, unlearning, and knowledge compression for deployment at the edge.
🔷
Brain-Inspired HD Computing
Hyperdimensional computing emulating human cognition via high-dimensional vector arithmetic — lightweight, error-tolerant, and neuro-symbolic AI for edge and IoT systems.
⚡
Alternative Computing & Systems
Processing-in-Memory (PIM), Near-Data Processing (NDP), multi-chip inference, and ML-driven system software for cross-platform performance prediction and resource optimization.

Project Pages