Seminar (SEM)

• Seminar in CL: User-facing language technologies: Practical implementation of high-quality NLP methods aimed at making the tools available to speakers.
• Above class meets with CSCI-B659

User-facing language technologies: Practical implementation of high-quality NLP methods aimed at making the tools available to speakers

Seminar (SEM)

• Topics in CL: Adv. Data modeling for CL

Topics in CL: Adv. Data modeling for CL Learn in depth about features of the Python programming language that are not covered in introductory courses, get experience in data modelling and program design, and algorithm implementation and evaluation. In this course you will gain experience in programming, and in implementing and evaluating research projects. Important skills you will learn are codebase organisation and understanding code written by others. Topics include: object-oriented programming, code archaeology and complexity analysis.

Seminar (SEM)

• Topics in CL: Adv. Machine Learning for CL
• Above class meets with CSCI-B 659

Advanced Machine Learning for CL focuses on new methods in CL based on Generative AI, Graph representations and Knowledge Graphs for semantic modeling, and Graph Neural networks. The topics discussed include also tuning of Large Language Models and LLMs with RAGs.

Seminar (SEM)

• Seminar in CL: Large Language Models
• Above class meets with CSCI-B 659
• basic programming proficiency in Python will be assumed

Large language models (LLMs) have dominated the attention of researchers in natural language processing and computational linguistics since the landmark releases of BERT (2018) and ChatGPT (2022). In this seminar, we survey work from the past two years on LLMs, covering their foundations, applications, interpretation, and ethical and legal considerations. After some preliminary assignments, students will spend the bulk of their time working on a project of their choosing related to LLMs. This course is designed primarily for graduate students in Computational Linguistics and Computer Science, but others are welcome. No technical background other than basic programming proficiency in Python is assumed, but students will benefit from already having some knowledge of deep learning and natural language processing.

Seminar (SEM)

• Topics in CL: Fundamentals of Speech Processing
• Above class meets with CSCI-B 590

An introduction to acoustic signal processing, focusing on the representation and analysis of continuous and discrete-time speech signals. Topics include phasors, sampling, FIR filters, the discrete-time Fourier transform, spectrum, and spectrogram analysis.