活动预告 | CodeWisdom 软件智能化开发与运维学术报告系列 第5期（高祥 北京航空航天大学）...

高祥

内容简介

题目

Improve the Correctness of Auto-Generated Programs

摘要

Program synthesis/generation attempts to infer a program according to a given specification. First, the specification can be demonstrated using concrete examples. However, the examples, which are usually represented in a form of input-output relations, can only specify part of the behaviors of the expected program. Therefore, test-driven program synthesis is prone to synthesize overfitting programs. Second, the specification can be given in the form of natural language descriptions. Large language models such as Codex, have shown the capability to produce code according to descriptions. However, language models lack awareness of program semantics, resulting in many incorrect programs.

I will present a series of approaches to improve the correctness of programs that are automatically generated by program synthesis/generation techniques. Our approaches are united by the idea of using program analysis to strengthen the specifications demonstrated via examples or descriptions.

报告人

高祥，北京航空航天大学副教授，新加坡国立大学获得博士学位。主要研究方向为程序分析、软件安全与程序合成。在CCF-A类软件工程和程序语言国际顶级期刊和会议（TOSEM、PLDI、OOPSLA、ICSE、ASE、ISSTA等）发表论文10余篇。获得新加坡国立大学研究成就奖，SIGSOFT杰出原型系统奖。所开发的技术已成功应用于微软智能IDE Visual Studio中，有效帮助开发人员提升开发效率。在软件工程顶级期刊TSE，TOSEM，ESE，TDSC等担任审稿人, JSS（Journal of Systems and Software）Open Science Board。

时间安排

❖

时间：2022/11/14（周一） 

          21:30-23:00

❖

地点：腾讯会议（397-865-610）