Subword Modeling

The goal of this course is to lead students to engage broadly with the existing NLP and computational linguistics research on subword modeling and develop new computational approaches to problems in morphology, orthography, and phonology. In addition to three other miniprojects, students will be expected to produce one piece of research that can be developed into a conference or workshop paper (though submission is not a course requirement). The paper should be suitable for the “Phonology, Morphology, and Word Segmentation” tracks of the *ACL conferences, the SIGMORPHON workshop, Coling, or LREC.

For a complete overview, see the syllabus.

Semester: Spring 2026

Location: WEH 5415

Day/Time: Tue and Thurs, 12:30 am — 1:50 pm

Contents:

• Schedule
• Instructors
• Course Objectives
• Prerequisites
• Grading
• Course Policies
• Student Resources

Course Schedule Top

In-Class Exercises Top

• Totonac Word-and-Paradigm Excercise with unpublished data from Jonathan Amith and Osbel Lopez. (For any use outside of this course, they should be attributed.)

Instructors Top

Instructor: Prof. David R. Mortensen

• Office Location: GHC 5407

• Office Hours: Tue 1:00–2:00pm, Wed 10:00–11:00am (Appointments always welcome)

• Email: dmortens@cs.cmu.edu
• Twitter: dmort27

Teaching Assistant: Rithvik Senthil

• Office Hours Location: GHC 5th floor common area (Near Pausch bridge entrance)

• Office Hours: Mondays - 11:00 am-12 pm, Thursdays - 12:30 pm-1:30 pm

• Email: rithviks@andrew.cmu.edu

Course Objectives Top

At the end of this course, students will:

• Express a sophisticated understanding of the fundamentals of morphology, phonetics, writing systems, and phonology
• Understand current papers in computational phonology, morphology, and orthography
• Implement current subword models of language and carry out publication-quality experiments in this area
• Write publication-quality papers reporting computational research in computational linguistics

Course Principles and Classroom guidelines

This course will be based around discussion of readings, guided by handouts. There will be no slides and no recordings of lectures. There are three groundrules:

• All class members are expected to have read readings prior to the beginning of the lecture. In many cases, readings will be divided among class members. This will be made clear during previous class meetings.
• All class members have a way of taking notes on handouts (e.g., a tablet + stylus or a printed copy and a pencil/pen). Handouts will be provided at least 24 hours in advance of class. If there is enough demand, the instructor can print copies of the handout for student use.
• All class members will treat other members of the class with dignity and respect.

Prerequisites Top

The course is open to graduate students and advanced undergraduates who have completed:

• One prior course in NLP
• One prior course in ML (preferred, but not required)

Grading Top

Grades will be based on three components:

• Highlights from each lecture (10 points)
• Projects (60 points)
• Final presentation and paper (30 points)

Required Texts Top

There are two required textbooks (in additional to many freely available readings):

• Haspelmath, M., & Sims, A. (2013). Understanding morphology. Routledge.
• Matthews, P. H. (1991). Morphology. Cambridge University Press.

Resources/Support/How to Succeed in This Course Top

Office Hours

Scheduled for 2 hours weekly. One-on-one meetings with the instructor can always be scheduled using Calendly.

Office of Disability and Access Services

Many people have disabilities, including members of our own families. We see disabilities as deficits not in disabled people but in the institutions and societies that are structured such that they are disadvantaged. We wish to do our part to overcome this disparate treatment. If you have a disability (visible or invisible), please let us know as soon as possible (you don’t need to tell us the nature of the disability) and work with Disability Service to develop a set of accommodations which we can then approve. These might, for example, include lecture materials that are usable by people with visual disabilities, sign language interpretation, captioning, flexible due dates, etc.

Support from the Instructor

Finally, if you tell me that you are having trouble, I will not judge you or think less of you. You do not owe me an explanation of your health (physical or mental) or the health of your loved ones; but you are welcome to tell me and I will listen. Even if I can’t help you directly, it is likely that I know someone who can. If you need help or more information, please ask, and I will work with you.

Class and University Policies Top

Diversity, Inclusion, Anti-racism

Throughout human history, some people have been denied the rights and opportunities available to others on the basis of their race, gender, economic class, caste, ancestry, language community, age, religion, beliefs, political affiliation, and abilities (visible and invisible). A single course cannot undo the injustices of history, but we—as a teaching staff—are committed to fighting inequity and promoting inclusion. We encourage you to join us. If you feel that you, or those around you, have been treated unfairly based upon their identity (or perceived identity) by us, by other members of the teaching staff, or by other students in the course, we ask that you bring it to our attention so that we can address the wrongs (as well as pursing the approved University channels).

Policy on Religious Observances

Due dates can be revised for documented religious observances. Please make sure to communicate about religious observances as far in advance as possible (and no later than one week before the observance, per university policy) so that accommodations can be made.

Assignment Completion & Late Work

This course works best when everybody completes their work by the designated deadlines. This prevents cascading tardiness from overwhelming both students and teaching staff. However, sometimes there are situations that call for extensions. Following university policy, we cannot grant extensions for these situations. However, we encourage you to work with the Disability Resources Office to request accommodations appropriate to you situation.

Academic Integrity

Any cheating or plagiarism will be dealt with according to the University policies on academic integrity. In general, discussion of tools, concepts, and formalisms is acceptable collaboration and is encouraged. Misrepresenting the work of others as your own, however, is considered cheating. If you are found to have cheated on an assessment, you will lose credit for that assessment and will be reported to the Office of Community Responsibility. The consequences of AIVs can be severe, including suspension or expulsion from the University.

License

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License