Vision-Language Models Courses and Certifications

Roboflow

Testing VLMs on Real-World Problems - How Do They Compare?

Explore how different vision-language models (VLMs) perform on real-world visual tasks, with comparisons, observations about their future, and techniques for combining models to solve complex problems.

• YouTube
• 22 minutes
• On-Demand
• Free Video

Center for Language & Speech Processing(CLSP), JHU

Introduction to Multimodal Large Language Models I - Day 10 Morning

Discover how multimodal large language models integrate text, audio, and visual data through expert-led tutorial slides from leading researchers at JSALT 2025.

• YouTube
• 2 hours 40 minutes
• On-Demand
• Free Video

Venelin Valkov

OCRFlux-3B - Local OCR AI Model Test - Turn PDFs into Markdown

Explore OCRFlux-3B, a fine-tuned Qwen2.5-VL model that converts document images to markdown, outperforming competitors on benchmarks through hands-on testing with financial statements and receipts.

• YouTube
• 11 minutes
• On-Demand
• Free Video

Montreal Robotics

Micro Vision Language Model Coding Tutorial

Discover how to build early-fusion Vision-Language Models from scratch, covering data processing, VIT creation from GPT code, multi-modal integration, and training optimization techniques.

• YouTube
• 1 hour 3 minutes
• On-Demand
• Free Video

USC Information Sciences Institute

Scale Can't Overcome Pragmatics - The Impact of Reporting Bias on Vision-Language Reasoning

Uncover how reporting bias in training data limits Vision-Language Models' reasoning capabilities, despite web-scale datasets, and explore intentional curation methods as solutions.

• YouTube
• 59 minutes
• On-Demand
• Free Video

Venelin Valkov

MinerU 2.5 - Local OCR VLM Text and Table Extraction Test

Explore MinerU 2.5, a 1.2B vision-language model for two-stage OCR with text, table, and formula recognition capabilities compared to classical OCR approaches.

• YouTube
• 12 minutes
• On-Demand
• Free Video

Discover AI

Failure of AI Visual Reasoning in Vision Language Models

Explore the limitations of AI visual reasoning in Vision Language Models through recent research findings and real-world experiences with commercial systems.

• YouTube
• 22 minutes
• On-Demand
• Free Video

Discover AI

BLIP-2: Connecting Vision-Language Models with Q-Former for Image Chat

Explore BLIP-2's innovative approach to combining vision and language models through Q-Former, enabling advanced image-based conversations and multimodal AI interactions.

• YouTube
• 13 minutes
• On-Demand
• Free Video

Montreal Robotics

RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control

Explore vision-language-action models for robotic control, enhancing generalization and semantic reasoning through Internet-scale data integration and end-to-end training.

• YouTube
• 1 hour 11 minutes
• On-Demand
• Free Video

OpenSource Connections

Vision Language Models and PDFs: What You See Is What You Search - Haystack EU 2024

Explore ColPali, a novel retrieval model using Vision Language Models to embed entire PDF pages, improving retrieval quality and simplifying extraction processes for complex document formats.

• YouTube
• 45 minutes
• On-Demand
• Free Video

Unlocking Insights From Multimodal PDFs Using OpenSearch and Vision-Language Models

Discover how to extract insights from multimodal PDFs using OpenSearch, OCR, ML models, and Vision-Language Models like ColPali for intelligent conversational search applications.

• YouTube
• 33 minutes
• On-Demand
• Free Video

Visual Quality Assurance Using Vision-Language Models

Explore how vision-language models revolutionize visual QA by detecting subtle UI failures and providing natural language explanations for smarter testing workflows.

• YouTube
• 51 minutes
• On-Demand
• Free Video

Discover AI

Fourier-Enhanced Fine-Tuning Vision Language Models Using PEFT-VFPT

Discover how Visual Fourier Prompt Tuning revolutionizes vision model fine-tuning by integrating Fast Fourier Transform, enhancing adaptability and accuracy while using minimal trainable parameters.

• YouTube
• 27 minutes
• On-Demand
• Free Video

Open Data Science

Mastering Visual AI with Vision-Language Models and Advanced Evaluation Techniques

Master computer vision for automotive damage detection using the CarDD dataset and FiftyOne platform to identify dents, scratches, cracks, and other vehicle damage types.

• YouTube
• 1 hour 8 minutes
• On-Demand
• Free Video

Broad Institute

Vision-Language Foundation Models for Cardiac Imaging

Explore EchoPrime, a multi-video vision-language model for comprehensive echocardiogram report generation, bridging AI and cardiology through advanced multimodal learning techniques.

• YouTube
• 37 minutes
• On-Demand
• Free Video