Reproduce “Co-Evolution of Types and Dependencies: Towards Repository-Level Type Inference for Python Code” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Shuo Sun, Shixin Zhang, Jiwei Yan, Jun Yan, Jian Zhang
PyTIR builds a repository-wide dependency graph that iteratively links Python entities needing type annotations, infers types using relevant context, and checks proposed annotations with Mypy. On 12 selected repositories, it reports TypeSim 0.89, compared with TIGER’s 0.70, and TypeExact 0.84. TypeSim measures behavioral similarity between predicted and developer-provided types. PyTIR targets repository-visible entities, not local variables or nested functions.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.