arrow
返回

How efficient is your clone? Evaluating LLM capabilities for runtime and memory-aware code selection

delete2026-06-23
delete0
delete
OA
AI
P
Prasanth Ramaswamy
M
Muhammad Azeem Akbar
M
Muzaffar Rao
A
Abdul Razzaq *
DOI:10.1016/j.jss.2026.112973delete
delete原文链接
delete分享
delete收藏
查看原文
摘要

摘要

En 中文
• 在CodeNet和CLBG上跨Python、Java和C的执行无关评估。 • Code+feature提示比编码器基线更能提升解码器LLMs。 • LoRA带来选择性增益;语言-方面效应仍依赖上下文。 • Top-1、MRR和eGap揭示了超越通常饱和的Top-3的权衡。
Keyword:
Large Language Models (LLMs)
Execution-free code selection
Functionally equivalent implementations
Runtime-aware code analysis
Memory-aware code analysis
Code efficiency
AI总结

AI总结

对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

Journal of Systems and Software 封面图
Journal of Systems and Software
IF:
4.1
论文数:
5.4K
被引数:
8.4K

机构

U
university of limerick
学者数:
1.2K
论文数: 580
被引数: 0
L
lut university
学者数:
263
论文数: 123
被引数: 2
引用论文

引用论文

err分享
err收藏
Comparison and evaluation of clone detection tools克隆检测工具的比较与评价
err2007-09-01
err510
errOAAI
errBellon, Stefan; Koschke, Rainer; Antoniol, Giuliano; Krinke, Jens; Merlo, Ettore
err分享
err收藏
CREF: An LLM-Based Conversational Software Repair Framework for Programming Tutors
err2024-09-11
err0
PREAI
errYang,Boyang; Tian,Haoye; Pian,Weiguo; Yu,Haoran; Wang,Haitao; Klein,Jacques; Bissyandé,Tegawendé F.; Jin,Shunfu
err分享
err收藏
Vulnerable Code Clone Detection for Operating System Through Correlation-Induced Learning
err2019-12-01
err10
PREAI
errShi, Heyuan; Wang, Runzhe; Fu, Ying; Jiang, Yu; Dong, Jian; Tang, Kun; Sun, Jiaguang
err分享
err收藏
学者 查看更多内容