Mechanistic Interpretability를 통한 LLM 내부 World Model 증명
Mechanistic Interpretability is a 2026 Breakthrough Technology. Here's What That Means for the "LLMs Are Just Matrix Multiplication" Debate
Feature Engineering에서 Representation Learning으로의 패러다임 전환
What Deep Learning Really Means — From Neural Networks to Modern AI
Information Bottleneck 설계를 통한 고차원 Semantic Representation 추출 전략
Autoencoders and Representation Learning in Vision
k-NN의 메모리 기반 검색 한계를 Linear Classifier의 학습 기반 예측으로 해결
Image Classification Explained — Why k-NN Breaks and Linear Classifiers Matter