y0news
← Feed
Back to feed
🧠 AI🟢 BullishImportance 6/10

Automating the Detection of Requirement Dependencies Using Large Language Models

arXiv – CS AI|Ikram Darif, Feifei Niu, Manel Abdellatif, Lionel C. Briand, Ramesh S., Arun Adiththan||6 views
🤖AI Summary

Researchers developed LEREDD, an LLM-based system that automates the detection of dependencies between software requirements using Retrieval-Augmented Generation and In-Context Learning. The system achieved 93% accuracy in classifying requirement dependencies, significantly outperforming existing baselines with relative gains of over 94% in F1 scores for specific dependency types.

Key Takeaways
  • LEREDD uses LLMs with RAG and ICL to automatically identify complex dependencies between natural language software requirements.
  • The system achieved 93% accuracy and 0.84 F1 score overall, with 0.96 F1 for non-dependent cases.
  • LEREDD showed 94.87% and 105.41% relative gains in F1 scores over baselines for Requires dependency detection.
  • Researchers created an annotated dataset of 813 requirement pairs across three systems to support future research.
  • The approach addresses a critical challenge in software development where requirement dependency detection is often overlooked or done manually.
Read Original →via arXiv – CS AI
Act on this with AI
Stay ahead of the market.
Connect your wallet to an AI agent. It reads balances, proposes swaps and bridges across 15 chains — you keep full control of your keys.
Connect Wallet to AI →How it works
Related Articles