Advanced topics in NLP (Master S3 Class)

This course will present some of the current trends in Natural Language Processing (NLP). Possible topics are:

the evolution of NLP systems,
common problems with NLP data (presence of biases, data contamination, etc.) and some of the challenges of system evaluation,
the strengths and weaknesses of large language models,
the explainability of NLP systems,
machine-learning beyond standard supervised learning in NLP (self-learning, reinforcement learning, unsupervised learning),
good practices for hyper-parameter search and model regularisation,
challenges and methods in low-resource settings (prototypical networks, data augmentation, etc.),
pragmatics and common-sense reasoning,
syntactic parsing and discourse analysis,
natural language generation,
the contributions of NLP tools to linguistics, including psycho-linguistics, language emergence and the link between distributional semantics and structuralism.

Knowledge in Natural Language Processing and Machine Learning.

On successful completion of this course, students should:

have a basic understanding of the historical evolution of Natural Language Processing,
understand the main challenges in Natural Language Processing today,
be able to critically read and present research articles on these topics.