New framework tackles ambiguity in natural language requirements

Researchers have developed a new framework to identify and resolve pragmatic ambiguities in natural language requirements using retrieval-augmented generation. This approach simulates stakeholders with varying domain expertise to detect interpretation discrepancies. The framework was evaluated on the PUblic REquirements dataset using GPT-4o-mini, Mistral-7B, Llama-3.1-8B, and Qwen2.5-7B models, showing promise in detecting ambiguities and generating clear, relevant disambiguated requirements. AI

IMPACT This framework could improve the clarity and accuracy of software development requirements, reducing misinterpretations and project delays.

RANK_REASON The cluster contains an academic paper detailing a new framework for detecting and resolving ambiguities in natural language requirements. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• arXiv
• GPT-4o mini
• Llama-3.1:8b
• mistral:7b
• Natural Language Requirements
• Pragmatic ambiguity and Kripke’s dialogue against Donnellan
• PUblic REquirements dataset
• qwen2.5:7b
• retrieval-augmented generation

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →