Reproduce “Learning to Search for Executable Patternmaking from Certified Programs” 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
Zhuotao Xu, Chi Zeng, Liying Yang, Xinxing Yu, Yanyan Liang
SemDraft recovers executable sewing-pattern edit programs from base and target patterns without recorded edit histories. It uses a deterministic compiler to certify sequences, then trains a learned prior to rank legal actions during bounded search. On 80 held-out synthetic cases, it certified 63 cases versus 51 for a heuristic within 1,000 search executions; results concern this generator and ontology.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.