paperbot · PL 论文追踪

全部论文 待读 收藏

P2 Programming not only by example

不仅通过示例编程
ICSE Proceedings of the 40th International Conference on Software Engineering2018匹配度 3/10 · 该论文属于交互式程序合成与开发者工具,虽处于 PL 与 AI 交叉领域,但并非用户关注的静态分析、LLM/Agent 改进静态分析或形式化验证、浮点程序分析、贝叶斯程序分析等核心方向。其“超越示例的意图表达”问题与用户研究画像仅有较边缘的关联。引用 37 inbox
Hila Peleg, Sharon Shoham, Eran Yahav
crossref: 2026-09-12

AI 速览

一句话:提出超越纯示例的交互式程序合成意图表达方式,允许程序员参照生成程序的具体部分来精化意图。

自动程序合成可根据程序员表达的意图生成代码,但意图传达仍是难题。粗粒度意图(如表达式期望类型的限制)常导致合成器返回大量候选,把筛选负担转移给用户。端用户合成中的示例编程(PBE)通过示例让用户交互式、迭代地精化意图,但仅依赖示例对程序员而言表达力不足:程序员能够观察生成程序,并直接针对生成程序的局部来精化意图。本文据此提出超越纯示例的意图表达与交互方式,让程序员把示例与生成程序的组成部分关联起来,以更精确地指导合成。摘要未给出具体实现与评估细节。

问题
自动程序合成的核心困难是传达程序员意图:粗粒度意图会产生大量候选,而仅用示例对程序员表达力不足。
方法
提出一种超越纯示例的交互式合成方式,让程序员观察生成程序并直接关联其局部来精化意图。
评估
摘要未给出具体评估设置或结果;论文主要提出并讨论该交互/意图表达方式。
原文摘要(Abstract)

Recent years have seen great progress in automated synthesis techniques that can automatically generate code based on some intent expressed by the programmer, but communicating this intent remains a major challenge. When the expressed intent is coarse-grained (for example, restriction on the expected type of an expression), the synthesizer often produces a long list of results for the programmer to choose from, shifting the heavy-lifting to the user. An alternative approach, successfully used in end-user synthesis, is programming by example (PBE), where the user leverages examples to interactively and iteratively refine the intent. However, using only examples is not expressive enough for programmers, who can observe the generated program and refine the intent by directly relating to parts of the generated program.

链接与引用

DOI 原文 · PDF(开放获取) ·

BibTeX
@inproceedings{paperbot4395,
  title = {Programming not only by example},
  author = {Hila Peleg and Sharon Shoham and Eran Yahav},
  booktitle = {ICSE},
  year = {2018},
  doi = {10.1145/3180155.3180189}
}