Reproduce “Same Problem, Different Field: Cross-Domain Solution Import via Domain-Stripped Computational Fingerprints” 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
Eryk Kulikowski
The authors introduce computational fingerprints for finding cross-field papers with similar computations. An LLM describes each paper’s computation while removing field vocabulary and method names; retrieval compares these descriptions. On a curated benchmark of known-swappable families, TF-IDF average precision rises from 0.222 for abstracts to 0.513 for mechanism skeletons and 0.557 for full fingerprints; matches remain unverified import candidates.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.