Grammatical Error Correction

Course Description

Grammatical Error Correction (GEC) is a task of automatically detecting and correcting errors in different texts. It has a wide range of applications, from proof-reading to assist language learners and children with developmental language disorders.

This course is a project seminar for senior undergraduate students, and highlights learning by doing. In this course, we will show students how to build neural networks and guide them to master seq2seq framework.

Instructor: Meng Li

Time: Fridays, 10:00 - 15:00 (April 11, 25; May 9; June 13)

Room: April 11, 25 - 2.14.0.26/27; May 9 and June 13 - 2.14.0.09;

Course Management System: Moodle

Syllabus

• Day 1 (2025-04-11):
  • 10:15 - 12:00. Transformer
  • 13:00 - 15:00. A Primer on Grammatical Error Correction
• Day 2 (2025-04-25):
  • 10:15 - 12:00. PyTorch Tutorial
  • 13:00 - 15:00. DL Projects in the Real World
• Day 3 (2025-05-09):
  • 10:15 - 12:00. NLP Research: From Zero to Hero
  • 13:00 - 15:00. Project Discussion
• Day 4 (2024-06-13):
  • 10:15 - 12:00. Project Presentation
  • 13:00 - 15:00. Project Presentation

Requirement

Prerequisites. You are expected to be proficient in Python programming, and have essential understanding of neural networks.

Format. In the first two blocks, I will prepare the essential knowledge that you need to develop a project on grammatical error correction. In the third block, I will present how to do NLP research, and discuss software projects. In May and June, you are expected to implement your project individually. Your project should change some variables (architectures, representations, languages, etc.) compared with the baseline. In the fourth block, each student should present their project and explain the code. I will ask random questions to check if you really understand your code.

Grading

• For bachelor students in Computational Linguistics: this seminar belongs to “Methods of computational linguistics and subject hours/Methoden der Computerlinguistik und Ver-suchspersonenstunden (LIN-BS-070)”.
  • To complete this seminar, you are expected to complete a software project on grammatical error correction, and give a presentation to explain your code at the end of this course.
  • (Optional) To pass the LIN-BS-070 module examination, you need to complete a term paper related to one of two seminars. If you would like to write a term paper on grammatical error correction, please talk to me.
• For bachelor students in Linguistics: this seminar belongs to “Computational linguistics (LIN-BS-034)”.
  • To complete this seminar and pass the module, you are expected to complete a software project on grammatical error correction, and write a term paper on NLP.

Resource

• Alice’s Adventures in a differentiable wonderland An intuitive, well-motivated reader on neural networks and their math foundation.
• Harvard CS197: AI Research Experiences A friendly course book for AI beginners.

Contact

Feel free to contact Meng at meng.li (at) uni-potsdam.de for any questions.