Efficient Information Retrieval – Part II

Your RAG system works in principle, but when it comes to real data, the system reaches its limits? Diverse file formats, unstructured content, large volumes and complex search queries result in relevant information not being found or answers losing quality.

In Part II of our workshop series, you will learn how to develop reliable and robust RAG systems or improve existing systems.

The focus is on advanced methods for processing various data types (e.g. PDF, HTML, XML, images, tables) and techniques for optimising RAG systems. Using realistic sample data sets, you will apply the methods and transfer them to your specific use case.

Key topics:

· Chunking strategies for different document types

· Handling long documents

· Multi-query retrieval

· Fusion strategies and re-ranking

· HNSW for efficient approximate nearest-neighbour search

· Scoring functions like BW25

· Fundamentals of Graph RAG and building a knowledge graph from your own documents

· Evaluation of RAG systems (Precision@K, Recall@K, MRR, nDCG)

· Differences between text embeddings and multimodal embeddings

👉 Part II expands on the technical fundamentals covered in Part I and prepares you for optimisation through human feedback in Part III.
After the workshop, you will be able to measure and improve the quality and reliability of your RAG system.

Requirements:
Participation in Part I of the workshop series
or equivalent experience with programming and RAG systems.

Registration & participation

Registration is binding. If you are unable to attend, please cancel your ticket in good time to allow other participants to take your place.

Important Information

Photographs, audio recordings and films will be made during the event. By participating in the event, you agree that photos, audio and video recordings in which you are recognizable may be published as part of the public relations work of the AI Service Center Berlin-Brandenburg and the HPI.

The is a project of the Federal Ministry of Research, Technology and Space. Its aim is to lower the barriers for using AI in business and society.

Are you interested in our free workshops and other events that we offer? Then sign up for our newsletter.

Know what’s Happening Next — before everyone else does.

Get App

Host or Publisher KI-Servicezentrum Berlin-Brandenburg @ HPI