Course overview
This course aims to provide an in-depth exploration of large language models (LLMs), emphasising the advanced principles of their architecture, development, and deployment. Students will engage with the intricacies of LLM training and benchmarking while critically analysing their potential and limitations in various real-world applications. A significant focus will be placed on the ethical, legal, and societal implications of LLMs, ensuring that students not only develop technical proficiency but also cultivate a responsible approach to their use in industry and academia.
Course learning outcomes
- Understand and describe the core principles behind LLMs, including their architectures, data sources, training pipelines, and benchmarking methodologies
- Develop and apply techniques for the training, fine-tuning, and deployment of LLMs, addressing real-world computational challenges and performance evaluation
- Evaluate and design innovative applications of LLMs across different domains, considering their capabilities, limitations, and emerging trends
- Critically assess the ethical, legal, and societal implications of LLMs, and design strategies for mitigating biases and ensuring responsible deployment
- Critically assess the limitations, risks, and future trends of LLM technologies, and propose innovative solutions for enhancing their ethical, societal, and technical impacts