Minjun Kim

Minjun Kim

M.S./Ph.D. Student,
Data Mining Lab,
Dept. of CSE,
Seoul National University
Seoul, South Korea

About Me

I am a third-year M.S./Ph.D. student majoring in Computer Science and Engineering at Seoul National University, advised by Professor U Kang. I received my B.S. in Computer Science from KAIST.

My research engages deeply with data mining and machine learning, specializing in the study of graph neural networks and model compression.

Email : minjun.kim [at] snu.ac.kr (Preferred)
Address : 6th Floor, Bldg. 303, 1 Gwanak-ro, Gwanak-gu, Seoul, South Korea
External Links : [ Homepage | Google Scholar | LinkedIn | DBLP | Github | ORCID | CV (pdf) ]

✈️ [Jun. 2026] I will be attending KCC 2026 and KDD 2026 both in Jeju, South Korea 🇰🇷! Please feel free to reach me out 😊. (24th-26th, Jun. & 9th-13th, Aug.)
[May 2026] An honor to receive top-tier reviewer awards from ICML 2026 and IJCAI 2026!
[Apr. 2026] One paper on parameter sharing has been accepted to ACL 2026. ([C7])
✈️ [Apr. 2026] I will be presenting 'Prune-then-Quantize or Quantize-then-Prune?' paper at ICLR 2026 in Rio, Brazil 🇧🇷! ([C6])
[Mar. 2026] Happy to receive the SNU BK21 Star Student Researcher Award!
[Jan. 2026] One paper on joint model compression has been accepted to ICLR 2026. ([C6])
✈️ [Jan. 2026] I will be presenting 'LampQ' paper at AAAI 2026 in Singapore 🇸🇬! ([C5])
[Nov. 2025] 'SynQ' paper has been selected as a Qualcomm Innovation Fellowship 2025 Finalist! ([C1])

Seoul National University (Mar. 2024 - Current)
M.S./Ph.D. Student in Computer Science and Engineering
Advisor: Prof. U Kang
KAIST (Feb. 2019 - Feb. 2024)
B.S. in Computer Science
(Double Major: Business and Technology Management)

* Equal Contribution, † Corresponding Author, Conference, Journal

SharVeT: Similarity-aware Parameter Sharing with Vector-based Tuning for Efficient LLM Compression

Jeongin Yun*, Jaeri Lee*, Jongjin Kim, Minjun Kim, Jinho Song, and U Kang^†
ACL 2026 (The 64th Annual Meeting of the Association for Computational Linguistics), San Diego, CA
[ Paper | Code | BibTeX ]
Prune-then-Quantize or Quantize-then-Prune? Understanding the Impact of Compression Order in Joint Model Compression

Minjun Kim, Jaehyeon Choi, Hyunwoo Yang, Jongjin Kim, Jinho Song, and U Kang^†
ICLR 2026 (The Fourteenth International Conference on Learning Representations), Rio de Janeiro, Brazil
[ Paper | Code | Poster | OpenReview | ICLR Virtual | arXiv | BibTeX ]
LampQ: Towards Accurate Layer-wise Mixed Precision Quantization for Vision Transformers

Minjun Kim, Jaeri Lee, Jongjin Kim, Jeongin Yun, Yongmo Kwon, and U Kang^†
AAAI 2026 (The 40th Annual AAAI Conference on Artificial Intelligence), Singapore
[ Paper | Code | Poster | arXiv | BibTeX ]

Zero-shot Quantization: A Comprehensive Survey

Minjun Kim*, Jaehyeon Choi*, Jongkeun Lee, Wonjin Cho, and U Kang^†
IJCAI 2025 (The 34th International Joint Conference on Artificial Intelligence), Montréal, Canada
Survey Track, Oral Presentation

[ Paper | List | Slides | Poster | arXiv | BibTeX ]
Unifying Uniform and Binary-coding Quantization for Accurate Compression of Large Language Models

Seungcheol Park, Jeongin Bae, Beomseok Kwon, Minjun Kim, Byeongwook Kim, Se Jung Kwon, U Kang^†, and Dongsoo Lee
ACL 2025 (The 63rd Annual Meeting of the Association for Computational Linguistics), Vienna, Austria
[ Paper | Code | ACL Anthology | arXiv | BibTeX ]
AugWard: Augmentation-Aware Representation Learning for Accurate Graph Classification

Minjun Kim, Jaehyeon Choi, SeungJoo Lee, Jinhong Jung^†, and U Kang^†
PAKDD 2025 (The 29th Pacific-Asia Conference on Knowledge Discovery and Data Mining), Sydney, Australia
Oral Presentation

[ Paper | Code | Slides | arXiv | BibTeX ]
SynQ: Accurate Zero-shot Quantization by Synthesis-aware Fine-tuning

Minjun Kim, Jongjin Kim, and U Kang^†
ICLR 2025 (The Thirteenth International Conference on Learning Representations), Singapore
Finalist, Qualcomm Innovation Fellowship 2025

[ Paper | Code | Poster | OpenReview | ICLR Virtual | arXiv | BibTeX ]

Electronic Device, Method, and Non-Transitory Computer Readable Storage Medium for Performing Quantization for Artificial Intelligence Model
Minjun Kim, Jaeri Lee, Jongjin Kim, Jeongin Yun, Yongmo Kwon, Samsung Electronics, and U Kang

Filed 31th July, 2025
Method and Apparatus for Accurate Zero-Shot Quantization by Synthesis-Aware Fine-Tuning
Minjun Kim, Jongjin Kim, and U Kang

Filed 23th May, 2025
Graph Classification Method and Apparatus based on Augmentation-Aware Representation Learning
Minjun Kim, Jaehyeon Choi, SeungJoo Lee, Jinhong Jung, and U Kang

Filed 30th Oct., 2024

AI Cluster @ Samsung Electronics - System LSI Division (Oct. 2025 - Current)

"Mixture-of-Experts Model Compression for Multi-Agent Systems"

MoE Merging, Pruning, and Quantization for Multi-Agent Systems
AI Star Fellowship @ IITP (Jul. 2025 - Current)

"Model Compression and Acceleration of Multi-modal LLMs"

Quantization, Pruning, Weight Sharing, and MoE Merging of MLLMs
AI Platform @ Samsung Electronics - MX Division (May. 2024 - Current)

"Mixed Precision Quantization of Large Language Models for On-Device Execution"

Mixed Precision Quantization of LLMs and MLLMs
SW-StarLab @ IITP (Jun. 2023 - Current)

"Model Compression for Deep Neural Networks"

Few-shot and Zero-shot Quantization of Vision Task Models
AI Research Project @ Youlchon Foundation (Jun. 2023 - Current)

"Advancing Language Models Through Compression of Large Language Models"

Model Compression of LLMs