Natural Language Generation
Meetings: Wed 3:00 - 4:20pm in CS 401
Instructor: Prof. Karthik Narasimhan (karthikn@cs.princeton.edu)
TA: Mengying Pan (mengying@cs.princeton.edu)
Office hours:
Karthik: Wed 4:20 - 5:20pm
Mengying: Fri 2 - 4pm
Course Description:
Recent advances in deep learning have led to exciting developments in natural language processing, especially in areas like translation, question answering and sentiment analysis. However, generating accurate and fluent language remains a major challenge even today. This seminar will study the latest research in natural language generation and allow students to choose and work on a research project in this space. The project can focus on either new theoretical or algorithmic developments in language generation, applications of generation to tasks in NLP, or development of new evaluation metrics for generation systems.
There are no prerequisites for this seminar beyond COS 217, COS 226 and one of COS 324/424/485, but prior experience with NLP is definitely useful. Students may work in pairs IF the project can be split so that each student has their own semester-size piece of the project. We will spend the first three meetings of the semester surveying the latest results in the field, brainstorming ideas and developing project proposals. The remaining meetings will be used for project updates, formal student presentations, and discussions on how to do background research, prepare a presentation and write a final paper.
Useful resources:
- Jurafsky and Martin, Speech and Language Processing
- Goodfellow, Bengio, Courville, Deep Learning
- Yoav Goldberg, A Primer on Neural Network Models for Natural Language Processing
- PyTorch tutorials: https://pytorch.org/tutorials/