Utilizing Language Learning Models for Parallel Scientific Code Generation and Translation

Abstract:  Large language models (LLMs) have the potential to serve as an assistant to aid scientists with scientific discovery. One important area needed for this assistance is that of parallel, scientific code generation. Existing LLMs, such as GPT-x, StarCoder and Code Llama, are being used to generate parallel, scientific code with some limitations.

This talk will discuss some of the limitations and present our framework, called LASSI, to address some of the limitations. LASSI, an LLM-based Automated Self-correcting pipeline for generating parallel ScIentific codes, features a pipeline that automatically generates code, tests for compilation and checks execution. LASSI has been extended to refactor code to generate energy-efficient parallel code for a given target platform.

For more information about upcoming speakers please visit the TPC Seminar Series Webpage: https://tpc.dev/tpc-seminar-series/