Reproduce “ArchiveLens: A Generative Spatial Knowledge Framework for Semantic Reconstruction of Historical Maps” 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
Yanan Wu, Yalin Yang, Chengbin Deng
ArchiveLens turns historical Sanborn map content into building records, combining footprints, material and text attributes, and source-sheet provenance. It retrieves records through semantic search with geographic and temporal filters. In a 50-question test, hybrid ST-RAG found records in the top five for 47/50 questions (94%), versus 35/50 (70%) for dense-only search on 445 Chicago sheets; it does not establish longitudinal change.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.