Profile photo of Seojune Lee

Seojune Lee

ECE Undergrad at Seoul National University

I am a final-year undergraduate student majoring in Electrical and Computer Engineering at Seoul National University. My primary academic interest lies in bridging the gap between deep learning algorithms and computer architecture. Specifically, I have been exploring methods that enable efficient training and inference of large language models (LLMs) through hardware-aware design. Aside from deep learning research, I am also interested in linguistics, software engineering, and quantitative finance. You can find more information in this link.


Publications

QWHA: Quantization-Aware Walsh-Hadamard Adaptation for Parameter-Efficient Fine-Tuning on Large Language Models

Hyesung Jeon*, Seojune Lee*, BeomSeok Kang, Yulhwa Kim, and Jae-Joon Kim
(*: equal contribution)

In ICLR 2026.


Recent Posts

LUT Tensor Core

Aug. 5, 2025 · 7 min read · 🌐︎ en/ko

MLSys architecture paper review quantization
Solving Triton Puzzles

July 30, 2025 · 3 min read · 🌐︎ en

GPU
Consistency Models

Feb. 2, 2025 · 6 min read · 🌐︎ en

generative deep learning paper review
Optimizing CUDA Matrix Multiplication: From Tiling to Tensor Cores

Dec. 22, 2024 · 7 min read · 🌐︎ en/ko

CUDA GPU
More Posts →

Updates

Ctrl+K
Start typing to search...