BriefGPT.xyz
Feb, 2025
走向神经符号程序理解
Toward Neurosymbolic Program Comprehension
HTML
PDF
Alejandro Velasco, Aya Garryyeva, David N. Palacio, Antonio Mastropaolo, Denys Poshyvanyk
TL;DR
本研究旨在解决大语言模型(LLMs)向大代码模型(LCMs)扩展过程中所面临的信任度和可解释性等问题,挑战了越来越多模型参数并非始终最佳的发展方向。研究提出一种神经符号方法,结合深度学习技术与传统符号方法的优势,旨在建立首个神经符号程序理解框架,以帮助识别缺陷代码组件,具有良好的应用前景。
Abstract
Recent advancements in
Large Language Models
(LLMs) have paved the way for Large Code Models (LCMs), enabling automation in complex software engineering tasks, such as code generation, software testing, and
Program Comp
→