May 1 | 4 pm CET | 10 am GMT-4
PDFs – When a Thousand Words Are Worth More Than a Picture (or Table)
This talk explores the challenges of using PDFs in Retrieval-Augmented Generation (RAG) systems and how multimodal models can enhance text retrieval. Discover how to overcome PDFs semantic limitation and learn techniques to improve the quality of extracted data for vector databases.
06
Days
20
Hours
15
Minutes
Date
May 1, 2025
Time
4 pm CET | 10 am EST | 7.30 pm IST | 5.00 pm KSA
Location
Online – Zoom
Language
English
Duration
1 hour
Transform How Your Models Understand PDFs!
In this hands-on, insight-packed session, you’ll explore how multimodal models can go beyond basic parsing and reverse engineering—breaking down complex tables and figures into meaningful, question-ready chunks. You’ll learn how to craft better input for vector databases and bring robustness to your retrieval layer, the foundation of any Retrieval-Augmented Generation (RAG) system.
The Webinar will cover:
- Understanding the semantic bottlenecks of PDFs and their impact on retrieval
- Reviewing chunk formats for vector database ingestion
- Using multimodal models to decompose tables into structured text
- Experimenting with retrieval quality using different extraction techniques
- Applying the same decomposition approach to figures for better knowledge retrieval
Who should attend?
This talk is ideal for AI practitioners, data scientists, and engineers working with RAG systems and document processing.
During this interactive session, you’ll learn how to enhance retrieval by leveraging multimodal models, improving PDF parsing strategies, and structuring extracted knowledge for better AI-driven insights.
Speaker
Meet the Computer Science and AI expert who will share his knowledge and expertise with you during AI Learning Week.
Caio Benatti Moretti
Caio holds a PhD in Computer Science and has been acting as a data scientist both in academia and industry since 2014. Currently working as a Data Science Consultant at Xebia, he created SlackGPT and is particularly keen on neural networks in its many forms and applications. His enthusiasm even led him to make a neural network fit inside a business card. Caio has been giving seminars on how to empower businesses with LLMs from use cases to technical tooling. He is focused on how LLMs can augment human productivity and hence helping businesses to leverage novel technologies to achieve their goals.
AI Learning Week Agenda
This year’s Academy open week will be held under the banner of AI.
This is a unique opportunity to develop your AI competencies and enter the Xebia Academy world.
We have prepared two tracks:
- Base Track is for anyone who wants to use AI to the best effect for themselves and their organization.
- Tech Track is designed for professionals who wish to explore prompt engineering, AI coding, and more.
April 28 | BASE TRACK
AI Strategies for Leaders: Ethics, Governance, and Business Success
Steven van Duin
April 29 | TECH TRACK
Practical GenAI: Building LLM-powered Applications
Jeroen Overschie
April 30 | BASE TRACK
Prompt Engineering Made Easy: Quick Tips to Get More out of ChatGPT
Lysanne van Beek
May 1 | TECH TRACK
PDFs – When a Thousand Words Are Worth More Than a Picture (or Table)
Caio Benatti Moretti
Reserve your spot
Sign up for the PDFs in the RAG Webinar. Remember, if you are interested in other AI Leadership Week webinars, you should register for each separately.
