ParaCoder: Parallel Code Generation with Large Language Model

High-performance parallel code generation is a complex and fascinating area in computer science that focuses on producing code that executes as quickly and efficiently as possible. In our paper, we designed a new architecture for parallel code generation agent with 4 inter-connected components of LLM---Memory, Planning, Tools and Action. It also incooperated with two techniques: data augmentation, prompting and retrieval-augmented editing to improve the performance of the parallel codes. Data augmentation is implemented by extracting and processing PIE dataset, and also synthesis dataset generated by LLM models with ParEval benchmark. Finally planning-oriented prompting, code verification and retrieval augmented editing are used to promote the actual performance of the LLM generated code. The evaluation results confirm that a rough speedup of 6.06X and 5.13X are achieved using Qwen2.5-Coder-7B-Instruct, Qwen2.5-Coder-14B-Instruct LLM models.

Authors

Institutions

Publication Details

Published
2025-03-01
DOI
https://doi.org/10.1145/3711708.3723442
Citations
1
Primary Topic
Parallel Computing and Optimization Techniques
Type
article
Field-Weighted Citation Impact
2.27
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

ParaCoder: Parallel Code Generation with Large Language Model

Lvfang Tao, Yanjie Wei, Jintao Meng, Minwen Deng et al.
1 citations
Parallel Computing and Optimization Techniques
2.27
article

ParaCoder: Parallel Code Generation with Large Language Model

Lvfang Tao, Yanjie Wei, Jintao Meng, Minwen Deng, Xu Zhang, Wenxi Zhu, Nan Zhou, Shengzhong Feng, Renjie Mao, Bingqiang Wang, Amelie Chi Zhou, Xiaowen Huang
article en
1 citations

Abstract

High-performance parallel code generation is a complex and fascinating area in computer science that focuses on producing code that executes as quickly and efficiently as possible. In our paper, we designed a new architecture for parallel code generation agent with 4 inter-connected components of LLM---Memory, Planning, Tools and Action. It also incooperated with two techniques: data augmentation, prompting and retrieval-augmented editing to improve the performance of the parallel codes. Data augmentation is implemented by extracting and processing PIE dataset, and also synthesis dataset generated by LLM models with ParEval benchmark. Finally planning-oriented prompting, code verification and retrieval augmented editing are used to promote the actual performance of the LLM generated code. The evaluation results confirm that a rough speedup of 6.06X and 5.13X are achieved using Qwen2.5-Coder-7B-Instruct, Qwen2.5-Coder-14B-Instruct LLM models.

Hong Kong Baptist University (HK), Shenzhen Institute of Information Technology (CN), Shenzhen University (CN), Shenzhen Polytechnic University (CN), Tencent (China) (CN), Southern University of Science and Technology (CN), Peng Cheng Laboratory (CN), Shenzhen Institutes of Advanced Technology (CN)
Openalex Percentile: Top 16%
Parallel Computing and Optimization Techniques
2.27
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.