Fall 2023
CSCI 699 · Ethics in NLP
Logistics
- Time:
- Tuesdays & Thursdays 4:00–5:50 PM
- Location:
- DMC 150
Teaching Team
Instructor: Jieyu Zhao
Office Hour: Thu 12–1 PM @ PHE 332
Teaching Assistants
- Pei Zhou — Mon 3–4 PM @ RTH 313
- Peifeng Wang — Thu 2:30–3:30 PM @ SAL 213
Introduction
Although there have been impressive advancements in natural language processing (NLP), several studies reported that NLP models contain social biases. Even worse, the models run the risk of further amplifying the stereotypes and causing harms to people. As NLP technology continues to advance and be integrated into various domains such as healthcare, finance, marketing, and social media, it raises important ethical concerns that need to be addressed. In this course, students will critically examine the ethical implications of NLP, including issues related to bias, fairness, privacy, transparency, accountability, and social impact. Through discussions, case studies, and guest lectures, students will explore the ethical challenges associated with NLP and develop a deep understanding of the ethical considerations that arise when designing, implementing, and deploying NLP applications.
Students will get a broad understanding about possible issues in current NLP models and how current research has tried to alleviate those issues. This class will equip students with the ability to read and write critical reviews about research papers, and learn how to conduct research related to NLP fairness, interpretability and robustness.
News
- First day of class is August 22.
Syllabus
Calendar subject to change. All work (except final report) is due 11:59 PM PT on the listed date.
| Wk | Date | Topic | Readings | Due |
|---|---|---|---|---|
| 1 | Tue Aug 22 | Introduction | ||
| Thu Aug 24 | Introduction (cont'd) | |||
| 2 | Tue Aug 29 | Project Examples / Human Subjects Research | ||
| Thu Aug 31 | Social Biases in NLP | |||
| 3 | Tue Sep 5 | Bias Evaluation | ||
| Thu Sep 7 | Models vs morality | |||
| 4 | Tue Sep 12 | Harms of LLMs in downstream tasks | ||
| Thu Sep 14 | Bias Mitigation | Project Proposal | ||
| 5 | Tue Sep 19 | Guest Lecture: What happened in Industry Research (Sunipa Dev) | ||
| Thu Sep 21 | Harms of LLMs: Bias and Stereotype | |||
| 6 | Tue Sep 26 | Hate Speech / Bias-Stereotype-Harm summary | ||
| Thu Sep 28 | Midterm Presentation Workshop | |||
| 7 | Tue Oct 3 | Midterm Presentation Workshop | ||
| Thu Oct 5 | Guest Lecture: LLM memorization (Eric Wallace) | |||
| 8 | Tue Oct 10 | Privacy | ||
| Thu Oct 12 | Fall Recess — No class | |||
| 9 | Tue Oct 17 | Human-Centered AI | ||
| Thu Oct 19 | Guest Lecture: Human & AI Interaction (Weiyan Shi) / Bias in Dialogue | |||
| 10 | Tue Oct 24 | Guest Lecture: Human-Centered AI (Sherry Wu) / Microaggression | ||
| Thu Oct 26 | Multilingual Biases | Midterm Report | ||
| 11 | Tue Oct 31 | Multimodal biases | ||
| Thu Nov 2 | Reduce bias in language models | |||
| 12 | Tue Nov 7 | Guest Lecture: Interpretation (Hanjie Chen) | ||
| Thu Nov 9 | Alignment with humans | |||
| 13 | Tue Nov 14 | Trade-offs between different metrics | ||
| Thu Nov 16 | Summary and Final Project Presentations | |||
| 14 | Tue Nov 21 | Final Project Presentations | ||
| Thu Nov 23 | Thanksgiving — No class | |||
| 15 | Tue Nov 28 | Final Project Presentations | ||
| Thu Nov 30 | Final Project Presentations | |||
| 16 | Tue Dec 5 | No class | ||
| Thu Dec 7 | No class | Project Report |
Logistics
- Office hours: Check above. For special requests, contact the course staff individually.
- Assignments: Submit through the course Google Drive folder shared by the course staff.
- Discussions: We use Slack for general course-related questions and announcements. For individual matters, email directly (please put [CSCI 699] in the subject line) or come to office hours.
Prerequisites
- Familiarity with NLP and/or ML. Ideal pre/co-requisites are CSCI 544 (Applied NLP) or CSCI 567 (Machine Learning).
- We will not teach any NLP concepts/techniques in this class!
- Programming skills. We mainly use Python with PyTorch, but you may use other libraries for your final project.
Grading
Grades are based on attendance (10%), paper presentation (30%), and a course project (60%).
Attendance and Discussion (10%):
- In-person attendance
Paper Reading and Discussion (30%):
- Present paper, lead class discussion
- Sign up as reviewers, peer-review others’ presentations
Course Project (60%):
- Project Proposal (10%)
- Midterm Report (10%)
- Final Presentation (20%)
- Final Report (20%)
Late Days
You have 4 late days for any assignment. Each late day allows submission 24 hours later than the deadline. If working in a group, one late day means each member spends a late day.
Paper Presentation
Paper presentation helps students develop the skills to give research talks. Each student presents 2 papers, prepares slides, and leads discussion. Each week another student signs up as the feedback provider (reviewer).
Grading rubric: correctness of content (40%), clarity (20%), discussion (20%), slides & presentation skills (20%).
Final Project
The final project can be done individually or in groups of up to 3. Each group should finish one research project related to the class topics. A “deliverable” result is expected — the project should be self-contained and reproducible. Typical successful projects: (1) a novel and sound solution to an interesting research problem, (2) correct and meaningful comparisons among baselines and existing approaches, or (3) applying existing techniques to a new application. Negative results are not penalized as long as the proposed approach is thoroughly explored and justified.
Use the standard *ACL paper submission template for project reports.
- Project proposal (10%) — Due Week 4, ~2 pages.
- Midterm progress report (10%) — Due Week 10, ~4 pages.
- Final presentation (20%) — Last two weeks, 30-min presentation with 5-min Q&A.
- Final project report (20%) — 8 pages total (excluding references), NLP conference format.
Resources
Related courses:
- UW Linguistics 575 — Ethics in NLP
- Berkeley CS 294 — Fairness in Machine Learning
- CMU — Computational Ethics for NLP