About

Table of contents

  1. Course Description
  2. Learning Objectives
  3. Grading
  4. Texts
  5. Mental Health
  6. Language Model Use Policy

Course Description

As more people are coming face-to-face with Large Language Models (LLMs), it is becoming increasingly important to study how these technologies interact and impact the humans who use them. Human-centered NLP and LLMs have received lots of attention recently, but what constitutes human-centered? Aren’t all languages ‘human-centered’?

In this course, we will explore emerging topics in HCI and NLP research to uncover what it means for language technologies, specifically LLMs, to be human centered. We will start with foundational research on human-centered design and how this work has been integrated into model development and evaluation. We will learn how ideas in HCI and NLP are intersecting in new and interesting ways, and try our hands at developing some of our own novel language interactions.

Classes will be a mix of lectures and seminars, with many class activities planned. The class is research focused. We will spend most classes on lectures and in-person discussions around research papers. Students will be expected to read papers and share ideas. There will be no exams. Instead, the class will culminate in a research project focused on an emerging topic in HCI or NLP. Classes will not be recorded.

Learning Objectives

  • Students can explain the basic concepts of HCI and NLP research, including common methodologies, theories and findings.
  • Students can analyze and critique research papers at the intersection of HCI and NLP
  • Students can use practical tools for building and analyzing language interactions

Grading

There is no exam in this course.

  • Leading a class discussion: 15%
  • Class participation: 35%
    • Attendance: 10%
    • Reading reactions 15%
    • In-class assignments: 10%
  • Project: 50%
    • Research pitch on 2 possible ideas: 5%
    • Talk on the design of your project (building on your pitch): 5%
    • Proposal write up: 10%
    • Project check-in (update on your project and current progress): 5%
    • Final talk: 10%
    • Final paper reporting your project and findings: 15%

Texts

Assigned readings will be linked in the class schedule. There are no required texts. More details about the assignments are in the Assignments section.

Mental Health

Diminished mental health, including significant stress, mood changes, excessive worry, substance/alcohol abuse, or problems with eating and/or sleeping can interfere with optimal academic performance, social development, and emotional wellbeing. The University of Illinois offers a variety of confidential services including individual and group counseling, crisis intervention, psychiatric services, and specialized screenings at no additional cost. If you or someone you know experiences any of the above mental health concerns, it is strongly encouraged to contact or visit any of the University’s resources provided below. Getting help is a smart and courageous thing to do – for yourself and for those who care about you.

Counseling Center: 217-333-3704, 610 East John Street Champaign, IL 61820

McKinley Health Center: 217-333-2700, 1109 South Lincoln Avenue, Urbana, Illinois 61801

Language Model Use Policy

The goal of this course is to think critically about current computing research on improving our use of language with technology. Language models (LMs) and chat assistants (e.g., ChatGPT) are some of the technologies we will be discussing in class, including how we can use them to write and read more effectively. For that reason, we are allowing the use of LMs to assist you in completing assignments. However, the use of any LM comes with the following requirements:

  • The output of the LM cannot be the final output that you submit for the assignment.
  • You must disclose the use of the LM on the assignment.
  • When you discclose LM use, also include 2-3 sentences describing why you used the LM, what you used it for, what you found helpful about using the LM, and what were limitations you had to work around.

The structure and content of this course has been inspired by Eshwar Chandrasekharan’s Social Computing course, and Katharina Reinecke’s Computer Ethics course.