Reproduce “Multimodal Requirements Traceability Link Recovery in Industrial Settings” 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
Haoming Huang, Shinpei Hayashi, Kazuya Kohata, Myeongki Kim, Yuki Kurihara
The system extends requirements traceability link recovery to text and image elements by retrieving candidate pairs separately by modality, having a vision-language model classify each pair, and filtering positive decisions by confidence. Across test sets, it recovered 49 of 171 gold links (precision .338, recall .287, F1 .310), but many links remained unrecovered; comparison with the text-only baseline changes multiple components.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.