8 papers
Hallucination Detection for LLM-based Text-to-SQL Generation via Two-Stage Metamorphic Testing
Bo Yang, Yinfen Xia, Weisong Sun +1
In Text-to-SQL generation, large language models (LLMs) have shown strong generalization and adaptability. However, LLMs sometimes generate hallucinations, i.e.,unrealistic or illo…
UCRBench: Benchmarking LLMs on Use Case Recovery
Shuyuan Xiao, Yiran Zhang, Weisong Sun +3
Use cases are widely employed to specify functional requirements, yet existing benchmarks are scarce and face the risk of being misaligned with actual system behavior, similarly li…
Knowledge-Guided Multi-Agent Framework for Application-Level Software Code Generation
Qian Xiong, Bo Yang, Weisong Sun +4
Automated code generation driven by Large Lan- guage Models (LLMs) has enhanced development efficiency, yet generating complex application-level software code remains challenging.…
Requirements Development and Formalization for Reliable Code Generation: A Multi-Agent Vision
Xu Lu, Weisong Sun, Yiran Zhang +4
Automated code generation has long been considered the holy grail of software engineering. The emergence of Large Language Models (LLMs) has catalyzed a revolutionary breakthrough…
MAAD: Automate Software Architecture Design through Knowledge-Driven Multi-Agent Collaboration
Ruiyin Li, Yiran Zhang, Xiyu Zhou +5
Software architecture design is a critical, yet inherently complex and knowledge-intensive phase of software development. It requires deep domain expertise, development experience,…
iReDev: A Knowledge-Driven Multi-Agent Framework for Intelligent Requirements Development
Dongming Jin, Weisong Sun, Jiangping Huang +4
Requirements development is a critical phase as it is responsible for providing a clear understanding of what stakeholders need. It involves collaboration among stakeholders to ext…