Reproduce “Beyond Accuracy: Disentangling Decomposition from Typed-State Quality Assurance in Multi-Agent LLM Pipelines” 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
Maria Gabriela Alves Zuppardo, Vinicius Cardoso Garcia
The authors compare three LLM architectures for classifying requirements as functional or non-functional and assigning MoSCoW priorities: one combined call, two specialised calls, and a typed-state pipeline. Across 625 PROMISE NFR+ requirements, three locally hosted models, and English and machine-translated Portuguese, the two-call variant generally shows larger Macro F1 gains; typed state mainly enables earlier validation, without proving operational benefit.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.