Generative AI, and in particular Large Language Models (LLMs), has already changed how we work and study. But this is just the beginning, as it has the potential of assisting and perhaps eventually automating knowledge workers in all areas, from law, medicine, to teaching and mental health therapists. This course will focus on the general principles and the latest research on methodologies and tools that can be applied to all domains. This is a project-oriented course, where students will gain hands-on experience in either methodology research or applying the concepts to create useful assistants for a domain of their choice. Topics include: (1) growing LLMs' knowledge through a combination of manual supervised learning and self-learning, (2) stopping LLMs from hallucination by grounding them with external corpora of knowledge, which is necessary for handling new, live, private as well as long-tail data, (3) handling external data corpora in different domains including structured and unstructured data, (4) experimentation and evaluation of conversational assistants based on LLMs, (5) controlling LLMs to achieve tasks, (6) persuasive LLMs, (7) multilingual assistants, and (8) combining voice and graphical interfaces. Prerequisites: one of LINGUIST CS 180/CS 280, CS 124, CS 224N, CS 224S, CS 224U.
3-4 units · Letter or Credit/No Credit
Generative AI, and in particular Large Language Models (LLMs), has already changed how we work and study. But this is just the beginning, as it has the potential of assisting and perhaps eventually automating knowledge workers in all areas, from law, medicine, to teaching and mental health therapists. This course will focus on the general principles and the latest research on methodologies and tools that can be applied to all domains. This is a project-oriented course, where students will gain hands-on experience in either methodology research or applying the concepts to create useful assistants for a domain of their choice. Topics include: (1) growing LLMs' knowledge through a combination of manual supervised learning and self-learning, (2) stopping LLMs from hallucination by grounding them with external corpora of knowledge, which is necessary for handling new, live, private as well as long-tail data, (3) handling external data corpora in different domains including structured and unstructured data, (4) experimentation and evaluation of conversational assistants based on LLMs, (5) controlling LLMs to achieve tasks, (6) persuasive LLMs, (7) multilingual assistants, and (8) combining voice and graphical interfaces. Prerequisites: one of LINGUIST 180/280, CS 124, CS 224N, CS 224S, 224U.
Offered in Autumn 2025 at Stanford University.