Introduction

Large Language Models (LLMs) such as GPT have two main drawbacks: knowledge cut-off and limited accuracy. This workshop introduces a method to enhance LLMs for these specific needs - Retrieval Augmented Generation (RAG). The workshop also introduces multimodal RAG, which adds multimedia capability to RAG.

Workshop Outcome

By the end of the workshop, students are expected to:

1. Gain understanding of Retrieval-Augmented Generation (RAG).
2. Get familiarize with Optical Character Recognition (OCR) and the process of creating text & image-caption dataset.
3. Be able to perform vector search and embedding.
4. Successfully implement multimodal RAG.

Resources

Get started with the following resources.

Slides:

https://www.canva.com/design/DAGPNkIPyH4/v14f906JrYRveIM2LpBF9w/edit?ui=eyJEIjp7IlAiOnsiQiI6ZmFsc2V9fX0

Unlocking Multimodal Intelligence: Workshop on RAG Systems

Workshop manual:

🌟 Unlocking Multimodal Intelligence: Workshop on RAG Systems 🌟

Codes used in the workshop:

https://github.com/HKUGenAI/LMM_RAG_Workshop