CV

M.S.

Korea Advanced Institute of Science and Technology (KAIST)

Computer Science

GPA: 4.17 / 4.30

B.S.

Ewha Womans University

Brain and Cognitive Sciences & Life Sciences

GPA: 4.07 / 4.30 (Summa Cum Laude)

• Thesis: Differential Mechanism of Positive and Negative Valence Representation: A Focus on Amygdalar Circuitry

Research Intern

CAMAIL, Max Planck Institute for Intelligent Systems / ELLIS Institute Tübingen

Research on blockwise parallelizable linear recurrences for efficient sequence modeling.

• Advisor: T. Konstantin Rusch

Master’s Student Researcher

Big Data Analysis & Learning Laboratory (BigDyL), KAIST

Research on efficient sequence modeling, Mixture-of-Experts (MoE), Large Language Models (LLMs)

• Advisor: Noseong Park

Research Intern

Big Data Analysis & Learning Laboratory (BigDyL), Yonsei University

Research on graph representation learning and graph neural networks (GNN).

• Advisor: Noseong Park

Data Scientist

DNI Consulting

Data analysis and modeling for B2B marketing and customer segmentation.

Undergraduate Research Intern

Cell Physiology Laboratory, Seoul National University

Research on electrophysiology and PFC-involved systems consolidation.

• Advisor: Seokho Lee

Undergraduate Research Intern

Neurobiology Laboratory, Seoul National University

Molecular analysis on dopaminergic reward circuitry.

• Advisor: Bongkiun Kaang

Q-Delta: Beyond Key–Value Associative State Evolution

Sumin Park, Seojin Kim, Noseong Park

ICML 2026

STAR: Rethinking MoE Routing as Structure-Aware Subspace Learning

Sumin Park, Noseong Park

ICML 2026

How Many Experts Are Enough? Towards Optimal Semantic Specialization for Mixture-of-Experts

Sumin Park, Noseong Park

AAAI 2026

Are Graph Transformers Necessary? Efficient Long-Range Message Passing with Fractal Nodes in MPNNs

Jeongwhan Choi, Seungjun Park, Sumin Park, Sung-Bae Cho, Noseong Park

AAAI 2026 (Oral)

DARS: Robust Sparse Fine-Tuning with Regularized Subspace Disalignment

Sumin Park, Noseong Park

ICLR 2025 Workshop (SCOPE)

PANDA: Expanded Width-Aware Message Passing Beyond Rewiring

Jeongwhan Choi, Sumin Park, Hyowon Wi, Sung-Bae Cho, Noseong Park

ICML 2024

Simulation-Based Defect Variable Identification from High-Resolution Images

Samsung–KAIST industry-academia project on simulation-based defect variable prediction using high-resolution images.

• Multi-task, multi-class classification framework
• ViT with hierarchical attention for high-resolution processing
• Achieved 99.41% classification accuracy

Summa Cum Laude

Ewha Womans University

Dean’s List (7 semesters)

Ewha Womans University

Academic Honors Scholarship

Ewha Womans University

Global Leader in Major

Ewha Womans University

Academic Honors Scholarship

Ewha Womans University

Admissions Scholarship (Full Tuition, 2 Years)

Ewha Womans University