Generic LLMs are great generalists, but they often fall short on tone or structure when you need them for a specific task. In this tutorial, we'll look at how to fine-tune an LLM to a specific task through additional training. We will discuss what fine-tuning actually does to a model, when it's better than just prompt engineering or other methods such as retrieval-augmented generation (RAG), and how to approach it. We'll walk through the full lifecycle, from framing the problem and preparing your data for fine-tuning, to training, evaluating, and shipping a specialized model. By the end, you'll have a solid grasp of how fine-tuning works and be ready to begin to apply it to your own tasks. Free and open to faculty, staff, and students, this session is led by a K-State graduate student.

Register to attend