Natural Language Processing — Fall 2026
Natural Language Processing
Foundations and modern methods for building computational systems that understand and generate human language.
Overview
About the course
This course provides an overview of both foundational and modern methodologies and tasks for the computational processing of human language, from probabilistic language models, sequence tagging, and syntactic parsing, to neural, transformer-based architectures that power emerging language assistants and AI agents like OpenAI’s ChatGPT, Google’s Gemini, and Claude Code. Homework assignments are designed to familiarize students with Python and PyTorch. Students have the opportunity to deepen their understanding of these concepts via a semester-long project.
Learning objectives
By the end of this course, students will be able to:
- Explain foundational and modern NLP methods.
- Implement NLP models using Python and PyTorch.
- Evaluate language-processing systems using appropriate metrics.
- Analyze the capabilities and limitations of neural and transformer-based models.
- Design, execute, and communicate an NLP project.
Reading materials
The primary textbook for the course is:
- Speech and Language Processing (3rd ed. draft)
Dan Jurafsky and James H. Martin
Additional readings, including research papers, tutorials, and other online resources, may be assigned throughout the semester.
Materials
Schedule
Check back regularly for schedule updates and assignment due dates.
Tentative special topics: Weeks 11–14 may change based on student interests, project needs, and course pace.
| Week | Date(s) | Topic | Slides & readings | Due |
|---|---|---|---|---|
| 1Overview | Mon., Aug. 24 | Welcome and syllabus | — | — |
| Wed., Aug. 26 | NLP tasks and ML foundations | — | — | |
| Fri., Aug. 28 | Words and tokenization | SLP C2 (sections 2.1–2.4) | — | |
| 2Language models | Mon., Aug. 31 | N-gram language models | SLP C3 | — |
| Wed., Sept. 2 | Evaluation | — | — | |
| Fri., Sept. 4 | Neural networks | — | Project proposal (PM1)Due Sept. 4 | |
| 3 | Sept. 7–11 | Neural language models | — | — |
| 4 | Sept. 14–18 | POS tagging and parsing | — | Homework 1 |
| 5 | Sept. 21–25 | Encoder–decoder models and attention | — | — |
| 6 | Sept. 28–Oct. 2 | Machine translation | — | Homework 2 |
| 7 | Oct. 5–9 | Transformer language models | — | — |
| 8 | Oct. 12–16 | Mid-term project presentations | — | Baseline report & presentation (PM2) |
| — | Oct. 17–25 | Mid-term break | — | — |
| 9 | Oct. 26–30 | Training large language models | — | Homework 3 |
| 10 | Nov. 2–6 | LLM post-training | — | — |
| 11 | Nov. 9–13 | Applied semantics and retrieval | — | Homework 4 |
| 12 | Nov. 16–20 | Evaluation and interpretability | — | — |
| 13 | Mon., Nov. 23 | Responsible NLP | — | Homework 5 |
| — | Nov. 25–29 | Thanksgiving holiday | — | — |
| 14 | Nov. 30–Dec. 4 | Multilingual and low-resource NLP | — | — |
| 15 | Mon., Dec. 7 | Poster session - joint evening session with CSE60556 Large Language Model; pizza and beverages provided. The regular daytime class may be canceled; details TBA. | — | Poster (PM3)Due Dec. 7 |
| Wed., Dec. 9 | Final report working session | — | — | |
| Thu., Dec. 10 | Reading days begin | — | — | |
| 16 | Dec. 14–17 | Final exam week (No classes) | — | Final report (PM4)Due Dec. 14 |
Assessment
Grading
Unless otherwise indicated, all assignments are due at 11:59 p.m. Eastern Time.
Regular attendance and active participation are expected. This includes contributing to class discussions and activities, responding to questions, and engaging with others respectfully. If you anticipate an absence, please notify the instructor in advance.
Five homework assignments worth 30 points each combine conceptual questions with hands-on programming in Python and PyTorch. They are designed to reinforce the methods discussed in class.
The semester-long project provides an opportunity to investigate an NLP problem in depth through four components: proposal (20 points), baseline report & presentation (30 points), poster presentation (30 points), and final report (40 points).
Letter-grade scale
| Letter grade | Points | Letter grade | Points |
|---|---|---|---|
| A | 280–300 | C+ | 230–239 |
| A− | 270–279 | C | 220–229 |
| B+ | 260–269 | C− | 210–219 |
| B | 250–259 | D | 180–209 |
| B− | 240–249 | F | 0–179 |
Expectations
Course policies
Honor Code
Students in this course are expected to abide by the Academic Code of Honor Pledge: “As a member of the Notre Dame community, I will not participate in or tolerate academic dishonesty.”
The following table summarizes how you may work with other students and use resources, including print and online sources and generative AI tools:
| Resources | Solutions | |
|---|---|---|
| Consulting | Allowed | Not Allowed |
| Copying | Cite | Not Allowed |
See the CSE Guide to the Honor Code for definitions of the above terms.
If an instructor sees behavior that is, in their judgement, academically dishonest, they are required to file either an Honor Code Violation Report or a formal report to the College of Engineering Honesty Committee.
Late Submission Policy & 24-Hour Extensions
Deadlines help keep the course moving smoothly, but occasional delays happen. Late homework will be handled as follows:
- 12-hour grace period: Every homework assignment has an automatic 12-hour grace period after the posted deadline without penalty.
- 24-hour extensions: You have three 24-hour extensions to use across homework assignments during the semester. To use an extension, email the instructor before the grace period expires. The instructor will confirm the extension and your remaining balance. Multiple extensions may be used on one assignment.
- Late penalty: After the grace period and any applied extensions expire, late work receives a 10% deduction for each additional 24-hour period or portion thereof.
- Cutoff: Work will not be accepted more than three days after the adjusted deadline.
Extensions apply only to homework, not to project milestones or presentations. University-approved absences are handled separately and do not use these extensions; notify the instructor by email as soon as possible.
Copyright
All course materials written by the instructor and published on this website are licensed under a Creative Commons Attribution 4.0 International License.
All other course materials, including lecture recordings and materials written by the instructor and distributed privately, should not be redistributed in any way; doing so is a violation of both US copyright law and the University of Notre Dame Honor Code.
Accessibility and Accommodations
Any student who has a documented disability and is registered with Sara Bea Accessibility Services should speak with the professor as soon as possible regarding accommodations. Students who are not registered should contact the Sara Bea Accessibility Services as soon as possible since accommodation typically needs to be arranged well in advance.
Student Mental Health & Well-Being
Diminished mental health can interfere with optimal academic performance. The source of symptoms might be related to your course work; if so, please speak with me. However, problems with other parts of your life can also contribute to decreased academic performance. The University Counseling Center (UCC) provides cost-free and confidential mental health services to help you manage personal challenges that threaten your emotional or academic well-being.
Remember, getting help is a smart and courageous thing to do — for yourself and for those who care about you. For more resources please see ucc.nd.edu or studenthealth.nd.edu.
The UCC is located on the third floor of Saint Liam Hall. Phone: 574-631-7336. Hours: Monday-Friday 8:00 a.m. – 5:00 p.m. Urgent Crisis Line 24/7
Title IX & Support Resources
The University of Notre Dame provides services for those who have been affected by sexual assault, sexual misconduct, dating or domestic violence, stalking and any conduct that creates a hostile environment. For help and further information including contact information for on and off-campus resources, please consult equity.nd.edu/resources.