paperbot · PL 论文追踪

全部论文 待读 收藏

P2 Proving hypersafety compositionally

组合式证明超安全性
OOPSLA 6(OOPSLA2)2022匹配度 2/10 · 论文属于程序验证/关系程序逻辑理论,聚焦 hypersafety 的组合性证明,与用户关注的 LLM/Agent 与静态分析/形式化验证交叉、浮点程序分析或贝叶斯程序分析均无直接关联;仅可作为形式化验证方向的背景知识。引用 15 inbox
Emanuele D’Osualdo, Azadeh Farzan, Derek Dreyer
crossref: 2026-09-10

AI 速览

一句话:提出 LHC 逻辑,以超三元组作为统一组合构件,实现更强的 hypersafety 证明组合性。

高阶安全性(hypersafety)性质刻画一个程序或一组程序的多条执行迹之间的关系,如确定性、幂等性与结合性。已有关系程序逻辑利用被关联程序的结构相似性来简化证明,但组合性受限。本文提出一种新的互补证明原则,将“超三元组”(hyper-triple)作为统一的组合式证明构件,并据此设计超三元组组合逻辑 LHC。LHC 支持以往逻辑难以实现的证明组合方式;作者证明其可靠性,并在若干有挑战性的例子中应用,展示其表达能力与实用性。

问题
如何以组合式方式证明元数为 n 的 hypersafety 性质,并克服已有关系程序逻辑在证明组合性上的限制。
方法
提出将 hyper-triple 作为统一组合构件的互补证明原则,并设计 Logic for Hyper-triple Composition(LHC)来支持新的证明组合形式。
评估
作者证明 LHC 的可靠性,并将其应用于若干具有挑战性的示例。
原文摘要(Abstract)

Hypersafety properties of arity n are program properties that relate n traces of a program (or, more generally, traces of n programs). Classic examples include determinism, idempotence, and associativity. A number of relational program logics have been introduced to target this class of properties. Their aim is to construct simpler proofs by capitalizing on structural similarities between the n related programs. We propose an unexplored, complementary proof principle that establishes hyper-triples (i.e. hypersafety judgments) as a unifying compositional building block for proofs, and we use it to develop a Logic for Hyper-triple Composition (LHC), which supports forms of proof compositionality that were not achievable in previous logics. We prove LHC sound and apply it to a number of challenging examples.

链接与引用

DOI 原文 · PDF(开放获取) ·

BibTeX
@article{paperbot1461,
  title = {Proving hypersafety compositionally},
  author = {Emanuele D’Osualdo and Azadeh Farzan and Derek Dreyer},
  journal = {Proceedings of the ACM on Programming Languages},
  volume = {6},
  number = {OOPSLA2},
  year = {2022},
  doi = {10.1145/3563298}
}