Scholar
Woomin Song
Google Scholar ID: fK24wGkAAAAJ
Ph.D Student, KAIST
Citations & Impact
Citations
53
H-index
2
i10-index
2
Publications
7
Co-authors
21
list available
Publications
Resume (English only)
Academic Achievements
- Published several papers including 'Compress, Gather, and Recompute: REFORMing Long-Context Processing in Transformers', 'Accelerated Test-Time Scaling with Model-Free Speculative Sampling', etc., presented at top conferences such as NeurIPS, EMNLP, and ICLR.
Research Experience
- Formerly an Applied Scientist Intern at Amazon AGI.
Education
- Received B.S. in Electrical Engineering / Computer Science (double major) / Mathematics (minor) from KAIST in 2022; currently a Ph.D. student at KAIST, advised by Prof. Jinwoo Shin.
Background
- Ph.D. student at KAIST AI, focusing on building efficient machine learning systems, particularly interested in reducing the inference cost of large language models (LLMs). Currently working on architectural modification techniques to improve computational efficiency of pre-trained models.
