Large Language Models (LLMs) are increasingly integrated into software engineering (SE) workflows, supporting tasks such as code generation, test case derivation, and requirements-to-code traceability. These tasks heavily rely on natural language requirements, making the quality of those requirements a critical factor in LLM performance. Previous research suggests that requirements smells, i.e. indicators of potential quality issues, can negatively affect the accuracy and reproducibility of LLM-generated outputs. However, the extent and nature of this impact remain largely unexplored. The dissertation aims to investigate how requirements smells influence LLM effectiveness across various SE tasks and to explore automated techniques for detecting and mitigating issues indicated by such smells. Initial results show that increasing the number of smells in requirements significantly degrades LLM performance in traceability tasks. The dissertation aims to establish empirical foundations for improving prompt quality and to support more reliable use of LLMs in SE through automated quality assurance mechanisms.

Requirements elicitation interviews are crucial for gathering system requirements but heavily depend on skilled analysts, making them resource-intensive, susceptible to human biases, and prone to miscommunication. Recent advancements in Large Language Models present new opportunities for automating parts of this process. This study introduces LLMREI, a chat bot designed to conduct requirements elicitation interviews with minimal human intervention, aiming to reduce common interviewer errors and improve the scalability of requirements elicitation. We explored two main approaches, zero-shot prompting and least-to-most prompting, to optimize LLMREI for requirements elicitation and evaluated its performance in 33 simulated stakeholder interviews. A third approach, fine-tuning, was initially considered but abandoned due to poor performance in preliminary trials. Our study assesses the chat bot’s effectiveness in three key areas: minimizing common interview errors, extracting relevant requirements, and adapting its questioning based on interview context and user responses. Our findings indicate that LLMREI makes a similar number of errors compared to human interviewers, is capable of extracting a large portion of requirements, and demonstrates a notable ability to generate highly context-dependent questions. We envision the greatest benefit of LLMREI in automating interviews with a large number of stakeholders.

Datasets containing human or animal movement are often sparse, for instance, due to poor GPS reception during recording or because of considerations concerning the battery life of the tracking device. Still, it may be desirable to estimate which walk the observed subject could have taken. A practical solution to this problem is to interpolate between measurements using random walk models. It is however intrinsically difficult to generate walks according to such models that also match the measurements, which in turn makes it difficult to compute metrics like visit probabilities.