"From Convolutions to Attention: Unpacking the Architecture Behind Modern AI"

May 15, 2025 · 2 min read · Introduction Transformers have revolutionized machine learning — powering models like BERT, GPT, T5, and even recent image models like ViT. But where did it all start? This blog breaks down the 2017 paper “Attention Is All You Need” by Vaswani et al....