Multimodal Large Language Models Theory Applications

This course introduces students to multimodal large language models that involve image, video, audio, speech, and text. You will be exposed to the mathematical working principles and architectures of recently proposed models and be involved in research methodology. In line with developing authentic learning, you will not only be trained on theoretical frameworks, but also hands-on coding activities and translating models to real-life applications via a service-learning component. Students who are keen on research, wish to develop creativity through designing their own multimodal large language models, and create project proposals to pitch their proposed model should take this course. This would assist them not only on technical skills but also on soft skills such as effective communication and articulation of ideas.