Explore a groundbreaking 18-minute conference talk from PLDI 2024 that delves into the space-efficiency problem of gradual typing with parametric polymorphism. Learn how researchers from Kyoto University, National Institute of Informatics, and Tokyo Institute of Technology propose a solution by slightly relaxing parametricity. Discover the development of λCmp, a coercion calculus with mostly parametric polymorphism, and its space-efficient counterpart λSmp. Gain insights into the proof of space-efficient execution for λSmp programs and the type- and semantics-preserving translation from λCmp to λSmp. Understand the implications of this research for efficient implementations of gradual typing in programming languages.
Overview
Syllabus
[PLDI24] Space-Efficient Polymorphic Gradual Typing, Mostly Parametric
Taught by
ACM SIGPLAN