Reproduce “LLM-Based Repair of Static Nullability Errors” 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
Nima Karimipour, Pascal Joos, Michael Pradel, Martin Kellogg, Manu Sridharan
NullRepair repairs residual nullability errors left after annotation inference in 12 Java projects by combining static analysis with an LLM that uses safe project examples to guide patches. In a combined sequential evaluation, retaining only patches that compile, resolve target errors, and add no new unresolved nullability errors, it resolves 704 of 1,119 cases (63%), below two baselines. The evaluation uses NullAway and Annotator.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.