CSED 503, Spring 2026: NLP and LMs
Wed, 6:30-9:20 in CSE2 G01
We will have five project assignments each worth 20% of the total class grade. They will be posted here when ready, starting in week 2 or 3. We will also give up to 5% extra credit bonus for strong class particiopation.
- Project 1: Text Classification with Logistic Regression (Due Apr 21)
- Project 2: N-Gram Language Models (Likely Due May 4, TBC)
- Project 3: Lexical Semantics (Likely Due 5/15, TBC)
- Project 5: Transformers (Likely Due 5/29, TBC)
- Project 5: Natural Language Generation (Likely Due 6/8,TBC)
On ChatGPT, Copilot, and other AI assistants (adopted from Greg Durrett): Understanding the capabilities of these systems and their boundaries is a major focus of this class, and there’s no better way to do that than by using them!
- We strongly encourage you to use ChatGPT to understand concepts in AI and machine learning. You should see it as a another tool like web search that can supplement understanding of the course material.
- You are allowed to use ChatGPT and Copilot for programming assignments. However, usage of ChatGPT must be limited in the same way as usage of other resources like websites or other students. You should come up with the high-level skeleton of the solution yourself and use these tools primarily as coding assistants.
- You are permitted to use ChatGPT for conceptual questions on assignments, but discouraged from doing so. It will get some of these questions right and some of them wrong. These questions are meant to deepen your understanding of the course content. Heavily relying on ChatGPT for your answers will negatively impact your learning.
An example of a good question is, “Write a line of Python code to reshape a Pytorch tensor x of [batch size, seqlen, hidden dimension] to be a 2-dimensional tensor with the first two dimensions collapsed.” Similar invocation of Copilot will probably be useful as well. An example of a bad question would be to try to feed in a large chunk of the assignment code and copy-paste the problem specification from the assignment PDF. This is also much less likely to be useful, as it might be hard to spot subtle bugs. As a heuristic, it should be possible for you to explain what each line of your code is doing. If you have code in your solution that is only included because ChatGPT told you to put it there, then it is no longer your own work in the same way.
- J&M III: Speech and Language Processing (Dan Jurafsky and James H. Martin)
- Eis: Natural Language Processing (Jacob Eisenstein)
Lecture slides will be posted on this site before the relevant day. We will alert the class if any major changes are made to correct errors, etc after posting.
We are eager to provide necessary accommodations. For disability accommodations, please see the UW resources. For religious accommodations, please see the UW resources.
We recognize that our students come from varied backgrounds and can have widely-varying circumstances. If you have any unforeseen or extenuating circumstance that arise during the course, please do not hesitate to contact the instructor to discuss your situation. The sooner we are made aware, the more easily these situations can be resolved. Extenuating circumstances may include:
- Work-school balance
- Familial responsibilities
- Unexpected travel
- Paper deadlines
- ... or anything else beyond your control which may negatively impact your performance in the class