Explore the application of Large Language Models (LLMs) in compiler optimization through a comprehensive 59-minute video. Delve into a groundbreaking research that introduces a custom 7 billion parameter Transformer model designed to enhance compiler performance. Learn how this innovative model takes unoptimized LLVM assembly as input and generates a list of optimal compiler options to minimize code size. Gain insights into the potential of LLMs in revolutionizing compiler technology and improving code efficiency. Discover additional resources for further exploration, including related research papers, AI trends, and industry insights. Connect with the Unify community through various platforms to stay updated on the latest developments in AI and compiler optimization.
Overview
Syllabus
Large Language Models for Compiler Optimisation
Taught by
Unify