Papers by Quazi Ishtiaque Mahmud
AutoParLLM: GNN-guided Context Generation for Zero-Shot Code Parallelization using LLMs (2025.naacl-long)
Copied to clipboard
Quazi Ishtiaque Mahmud, Ali TehraniJamsaz, Hung D Phan, Le Chen, Mihai Capotă, Theodore L. Willke, Nesreen K. Ahmed, Ali Jannesari
| Challenge: | In-Context Learning (ICL) is a powerful technique to augment the capabilities of LLMs for a diverse range of tasks. |
| Approach: | They propose a way to generate context using guidance from graph neural networks to generate efficient parallel codes. |
| Outcome: | The proposed method improves state-of-the-art LLMs by 19.9% and 6.48% on NAS and rodinia benchmarks. |