About
I am a Ph.D. student in the Interdisciplinary Program in Artificial Intelligence at Seoul National University, advised by Prof. Sungjoo Yoo in the Computing and Memory Architecture Laboratory.
My research focuses on efficient multi-agent systems, LLM quantization, KV-cache quantization, LLM pruning, and diffusion LLMs. My work explores how to compress models and reuse computation for more efficient inference.
Research interests: Efficient multi-agent systems · LLM quantization · KV-cache quantization · LLM pruning · Diffusion LLMs
Publications
* Equal contribution
Prefill-Free Cross-Family KV Cache Transfer for Heterogeneous Multi-Agent LLMs
Output-Aware Rotation for INT2 KV-Cache Quantization
Locality-Aware Redundancy Pruning for LLM Depth Compression
Rethinking Layer Redundancy in Large Language Models: Calibration Objectives and Search for Depth Pruning
EntropyCache: Decoded Token Entropy Guided KV Caching for Diffusion Language Models
Grouped Sequency-arranged Rotation: Optimizing Rotation Transformation for Quantization for Free
Rotate, Clip, and Partition: Towards W2A4KV4 Quantization by Integrating Rotation and Learnable Non-uniform Quantizer
Education
MAR 2024 – PRESENT
Seoul National University
Ph.D. Student, Interdisciplinary Program in Artificial Intelligence
Advisor: Prof. Sungjoo Yoo
MAR 2017 – FEB 2024
Seoul National University
B.S. in Electrical and Computer Engineering
Experience
PRESENT
Graduate Researcher
Computing and Memory Architecture Laboratory
Seoul National University
SUMMER 2022
Intern · SK Hynix
CIS Process Integration · Icheon, South Korea
300 mm time-of-flight sensors and their core technologies.
Teaching
Teaching Assistant · Seoul National University
- Advanced Computer Architecture
- Computer Architecture
- Hardware System Design
Honors
2022 · SPRING & FALL
Merit-based Academic Scholarship
Seoul National University