28938

Domain-Specific Code Language Models: Unraveling the Potential for HPC Codes and Tasks

Tal Kadosh, Niranjan Hasabnis, Vy A. Vo, Nadav Schneider, Neva Krien, Mihai Capota, Abdul Wasay, Nesreen Ahmed, Ted Willke, Guy Tamir, Yuval Pinter, Timothy Mattson, Gal Oren
Ben-Gurion University, Israel
arXiv:2312.13322 [cs.PL], (20 Dec 2023)

@misc{kadosh2023domainspecific,

   title={Domain-Specific Code Language Models: Unraveling the Potential for HPC Codes and Tasks},

   author={Tal Kadosh and Niranjan Hasabnis and Vy A. Vo and Nadav Schneider and Neva Krien and Mihai Capota and Abdul Wasay and Nesreen Ahmed and Ted Willke and Guy Tamir and Yuval Pinter and Timothy Mattson and Gal Oren},

   year={2023},

   eprint={2312.13322},

   archivePrefix={arXiv},

   primaryClass={cs.PL}

}

Download Download (PDF)   View View   Source Source   Source codes Source codes

Package:

10978

views

With easier access to powerful compute resources, there is a growing trend in AI for software development to develop larger language models (LLMs) to address a variety of programming tasks. Even LLMs applied to tasks from the high-performance computing (HPC) domain are huge in size and demand expensive compute resources for training. This is partly because these LLMs for HPC tasks are obtained by finetuning existing LLMs that support several natural and/or programming languages. We found this design choice confusing – why do we need large LMs trained on natural languages and programming languages unrelated to HPC for HPC-specific tasks? In this line of work, we aim to question choices made by existing LLMs by developing smaller LMs for specific domains – we call them domain-specific LMs. Specifically, we start off with HPC as a domain and build an HPC-specific LM, named MonoCoder, that is orders of magnitude smaller than existing LMs but delivers similar, if not better performance, on non-HPC and HPC tasks. Specifically, we pre-trained MonoCoder on an HPC-specific dataset (named HPCorpus) of C and C++ programs mined from GitHub. We evaluated the performance of MonoCoder against conventional multi-lingual LLMs. Results demonstrate that MonoCoder, although much smaller than existing LMs, achieves similar results on normalized-perplexity tests and much better ones in CodeBLEU competence for high-performance and parallel code generations. Furthermore, fine-tuning the base model for the specific task of parallel code generation (OpenMP parallel for pragmas) demonstrates outstanding results compared to GPT, especially when local misleading semantics are removed by our novel pre-processor Tokompiler, showcasing the ability of domain-specific models to assist in HPC-relevant tasks.
No votes yet.
Please wait...

* * *

* * *

HGPU group © 2010-2024 hgpu.org

All rights belong to the respective authors

Contact us: