Reproduce “Semantic-Enhanced Automatic Refinement of Architecture Recovery Results Using LLMs” 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
Yiran Zhang, Chengwei Liu, Yuqiang Sun, Zhengzi Xu, Weisong Sun, Wenke Li, Wuxia Jin, Yang Liu
SemRef refines software architecture recovery results by combining dependency evidence with an LLM’s judgments about file and module roles, then organizing modules into a hierarchy. Across nine projects and 10 recovery tools, it improved all five reported similarity metrics; mean MoJoFM rose from 0.3935 to 0.6684. Evaluation compares flattened hierarchies with published human-labeled architectures, not uniquely correct structures.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.