Reproduce “Tuning LLM-based Code Optimization via Meta-Prompting: An Industrial Perspective” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Jingzhi Gong, Rafail Giavrimis, Paul Brookes, Vardan Voskanyan, Fan Wu, Mari Ashiga, Matthew Truscott, Mike Basios and 3 more
MPCO generates tailored code-optimization prompts from project, task, and target-model context in one inference call, then sends them to optimizer models. It is evaluated on five C++ and Python codebases against a fixed contextual-prompt baseline using the same fields. MPCO ranks best overall; on Llama.cpp, mean runtime improvement is 7.84% versus 3.97%. Evidence is limited to tested runtime tasks.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.