You built a Retrieval-Augmented Generation (RAG) system for…

You built a Retrieval-Augmented Generation (RAG) system for your company.Executives often ask broad, multi-part questions such as “What is our travel policy for contractors and interns?” The AI assistant usually retrieves relevant passages, but the coverage feels incomplete — each answer includes only part of the needed information. To help the system capture more complete answers without drastically increasing noise, what adjustment should you try first?

You’re preparing a dataset to fine-tune your company’s custo…

You’re preparing a dataset to fine-tune your company’s customer-service LLM. The goal is to make the model both accurate and generalizable across different types of customer requests. Which of the following are characteristics of high-quality Supervised Fine-Tuning (SFT) data?(Select all that apply.)

You are a project manager, leading the AI alignment project…

You are a project manager, leading the AI alignment project at Aurora Media, where your team is refining a brand-language model to generate emotionally engaging ad copy. Your data scientists have collected human preference labels comparing which ad responses sound more authentic.They are now deciding whether to apply Reinforcement Learning from Human Feedback (RLHF) or Direct Preference Optimization (DPO) for alignment. Which step is specific to RLHF and not part of DPO’s training procedure?

Many political scientists advocate something known as “respo…

Many political scientists advocate something known as “responsible parties” and the “responsible party model of government” (RPMG).  According to lecture and the text, which of the following best describes how members of Congress should behave if they follow the responsible party model?