Abstract
Many synthesizers implicitly benefit from using polymorphic types, since parametric polymorphism reduces the search space. Additional synthesis constraints may interfere with parametricity. In particular, a polymorphic type may cause otherwise feasible input-output examples to contradict each other. We present Taxi (type-and-example based inferencer), a tool for efficiently reasoning about the feasibility of polymorphic programs specified by input-output examples, and Driver, a tactic language for top-down program synthesis that uses feasibility reasoning to prune the search space. Taxi guarantees that every search state corresponds to a correct (albeit possibly partial) implementation. In addition, it allows for shortcutting the synthesis when a subspecification covers all cases. We show that these techniques have the potential to speed up top-down enumerative type-and-example driven synthesizers.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the 2026 ACM SIGPLAN International Workshop on Partial Evaluation and Program Manipulation (PEPM ’26) |
| Publisher | Association for Computing Machinery |
| Pages | 17-30 |
| Number of pages | 14 |
| ISBN (Print) | 979-8-4007-2357-5 |
| DOIs | |
| Publication status | Published - 8 Jan 2026 |
Bibliographical note
Publisher Copyright:© 2026 Owner/Author.
Keywords
- container functors
- example propagation
- feasibility
- parametricity
- program synthesis
Fingerprint
Dive into the research topics of 'Hole Refinements for Polymorphic Type-and-Example Driven Synthesis'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver