• question answering
• language model
• deep learning
• biomedical research
• generative search engine

This master thesis aims to develop an evidence-based question-answering system for biomedical research in collaboration with Bayer pharmaceutical. The research focuses on addressing challenges such as retrieving relevant information, customization, privacy, and efficiency. The proposed system leverages natural language processing techniques and large language models to enable users to ask questions in an intuitive manner and receive accurate, informative answers. The study explores the feasibility of constructing a generative search engine for scientific literature and investigates the potential of transferring the capabilities of a large language model to a smaller, more privacy-conscious model while maintaining comparable performance.