Human-Centred Natural Language Processing

Course Description

The Human-Centered Natural Language Processing (NLP) course is designed to provide an advanced understanding of how AI systems can process and generate human language. By focusing on the interplay between machine learning, linguistics, and human communication, this course explores how NLP technologies are developed to work alongside humans, amplifying their abilities rather than replacing them.

Course Objectives

This course is tailored for students with a basic background in Machine Learning and Python Programming, and aims to equip participants with the knowledge and tools to build AI systems that understand and generate language in ways that are useful, ethical, and aligned with human needs.

2. Key research and application areas

Natural Language Processing (NLP):

NLP is the field of AI dedicated to teaching machines to understand, interpret, and generate human language. It leverages deep learning techniques, linguistic models, and computational frameworks to enable machines to analyze text or speech, capturing nuances like sentiment, intent, and context. Topic in NLP include:

Tokenization & Representation: Breaking down text into units that machines can process while preserving meaning.
Language Models: Leveraging state-of-the-art models like Transformers to understand and generate human-like text.
Sequence-to-Sequence Models: Enabling machines to perform tasks like translation, summarization, and question-answering by predicting the next word or phrase in a sequence.

Ethical AI in NLP

The course emphasizes the ethical dimensions of AI in language technologies. Trustworthy AI in NLP ensures that systems are lawful, ethical, and robust. We will examine how NLP models can:

Respect Human Autonomy: Ensure users retain control over decisions made by AI.
Prevent Harm: Minimize biases and errors in language models that could lead to negative consequences.
Promote Fairness: Address issues like bias in training data, ensuring that AI-generated language is equitable and inclusive.
Ensure Explainability: Make AI decisions understandable to humans, enabling transparency and trust in model outcomes.

Generative AI & Large Language Models:

Generative AI plays a pivotal role in modern NLP. This section of the course covers how large language models (e.g., GPT) are trained to generate coherent and contextually relevant text. Topics include:

- Pre-Training & Fine-Tuning: Understanding how large models are trained on vast amounts of data, and then fine-tuned for specific tasks.
- Prompting: Techniques to guide the model in producing specific responses based on user inputs.
- Reinforcement Learning from Human Feedback: Enhancing models to align outputs with human values and expectations through iterative feedback loops.

Human-AI Collaboration:

The course takes a human-centered approach, focusing on how NLP systems can collaborate with humans in real-world applications. Students will explore how AI can assist in tasks such as:

Question-Answering Systems: AI systems that provide accurate, context-aware responses to human queries.
Retrieval-Augmented Generation (RAG): Combining information retrieval with generative models to produce more reliable, knowledge-rich responses.
Multimodal Language Models: Integration of text and visual data to enable systems like Visual Question Answering, where machines can interpret both language and images.

Explainability & Transparency in NLP:

Understanding Explainability in NLP models is crucial for creating AI systems that users can trust. This section will delve into methods for making deep learning models more interpretable:

Model Interpretability: Tools and techniques that provide insight into how NLP models make decisions.
Transparency in Decision-Making: Ensuring that users understand why a model produced a certain result, thereby fostering trust and reducing the risk of errors.

Syllabus

TBD

Useful Resources

Moodel Page, LSF Page (TBD), Notion Page (TBD)