LLM Lecture Series: How Large Language Models Work

LLM Lecture Series: How Large Language Models Work

The LLM Lecture Series: How Large Language Models Work is an advanced training programme organised by the Malta Digital Innovation Authority (MDIA) and DiHubMT, in collaboration with the University of Malta Faculty of ICT and the IEEE Malta Section. The series is designed to provide participants with a deeper technical understanding of the technologies underpinning modern AI systems such as ChatGPT, Claude, Gemini and other Large Language Models (LLMs).

Delivered over six sessions, the programme explores the foundations of artificial intelligence, machine learning and natural language processing before progressing into the architecture and training of Transformer-based models. Participants gain insight into concepts such as neural networks, embeddings, attention mechanisms, tokenisation, model training, fine-tuning, emergent properties and the practical challenges associated with modern generative AI systems.

The lecture series is aimed at ICT professionals, engineers, software developers, researchers, students and technically curious professionals seeking to understand not only how to use AI tools, but how they work internally. The programme combines theoretical foundations with practical explanations, including detailed breakdowns of Transformer architectures and the mechanisms that enable modern language models to generate human-like responses.

Through the collaboration between MDIA, DiHubMT, the University of Malta Faculty of ICT and the IEEE Malta Section, the initiative supports Malta’s growing AI ecosystem by promoting digital skills development, technical excellence and informed discussion around the opportunities and challenges presented by artificial intelligence.

Organisers and Partners

• Malta Digital Innovation Authority
• DiHubMT
• University of Malta Faculty of ICT
• IEEE Malta Section

Theme: Understanding the architecture, training and computational mechanisms behind Large Language Models and Generative AI.