Data-Driven Dynamic Algorithm Dispatch with Large Language Models

Abstract:We introduce a large language model (LLM)-driven approach for generating dynamic algorithmic dispatch heuristics in high-performance linear algebra. By combining prompt engineering with LLaMA 3 and a curated performance database, the model learns to synthesize selection heuristics that exploit structural patterns to identify fast algorithmic choices. A case study on LU factorization demonstrates the model's ability to replicate expert-designed strategies. This work, developed as part of the DARPA-MIT SmartSolve project, highlights the promise of LLMs for algorithmic discovery and the development of more adaptive, fast linear algebra software.

3 pages, 2 figures. Accepted at the 2025 IEEE High Performance Extreme Computing Conference (HPEC). Outstanding Short Paper Award

Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE); Numerical Analysis (math.NA)

arXiv:2608.21584 [cs.AI]

(or arXiv:2608.21584v1 [cs.AI] for this version)

https://doi.org/10.48550/arXiv.2608.21584 Focus to learn more arXiv-issued DOI via DataCite (pending registration)

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.