Бакалавриат
2026/2027
Обработка естественного языка
Статус:
Курс по выбору (Прикладной анализ данных)
Где читается:
Факультет компьютерных наук
Когда читается:
4-й курс, 1, 2 модуль
Охват аудитории:
для своего кампуса
Язык:
английский
Кредиты:
5
Контактные часы:
56
Course Syllabus
Abstract
The course “Natural Language Processing” introduces fundamental and modern methods for processing and analyzing natural language at the intersection of machine learning, deep learning, and computational linguistics. The course covers classical NLP methods, neural approaches to text processing, language modeling, attention and Transformer-based architectures, and modern large language models (LLMs). Particular attention is given to practical applications of LLMs, including fine-tuning, retrieval-augmented generation, efficient inference, evaluation, tool use and agents, as well as multimodal models and selected topics in reliable and responsible NLP.
Learning Objectives
- The course aims to introduce students to the fundamental concepts, methods, and applications of natural language processing, from classical text processing and neural NLP models to modern large language models and multimodal systems. Students will gain both theoretical understanding and practical experience with contemporary NLP tools, LLM-based applications, retrieval-augmented generation, agents, and vision-language models.
Expected Learning Outcomes
- Apply standard text preprocessing, representation, and analysis techniques to natural language data.
- Formulate common NLP tasks and select appropriate models and evaluation methods for solving them.
- Explain the main principles of language modeling, attention, Transformer-based architectures, and modern large language models.
- Fine-tune, adapt, and apply pretrained language models and LLMs to practical NLP tasks.
- Design and use modern NLP systems, including retrieval-augmented generation, structured generation, agents, and multimodal models.
- Evaluate NLP and LLM systems and identify key limitations related to reliability, bias, safety, and responsible use.
Course Contents
- Introduction to NLP and text representations
- Neural sequence models and language modeling
- BERT, GPT and transfer learning
- NLP tasks and fine-tuning
- LLM training and alignment
- Attention and Transformer
- Modern LLM architectures
- Information retrieval and RAG
- Prompting and efficient adaptation
- Efficient LLM inference
- LLM applications: structured generation, tools and agents
- LLM evaluation
- LLM applications: structured generation, tools and agents.
- Reliable and responsible NLP
- Multimodal and modern models
