InTowards AIbyMadjid Korrabi·Nov 11Understanding LLMs from Scratch — Part 2: From Tokens to Semantic VectorsA response icon1A response icon1
InTowards AIbyMadjid Korrabi·Nov 3Understanding LLMs from Scratch — Part 1: How LLMs Process Human Language (Tokenization)
Madjid Korrabi·Jun 24Mastering Sequential Data [2/2]: Introduction to Long Short-Term Memory (LSTM)In my previous article on Recurrent Neural Networks (RNNs), I discussed RNNs and how they work. Towards the end of the article, the…
Madjid Korrabi·Jun 21Mastering Sequential Data [1/2]: A Deep Dive into RNNsIf you’re interested in Artificial Intelligence — especially Natural Language Processing (NLP) — you’ve probably come across the term RNN…
Madjid Korrabi·Jun 19CNN Part 2: No More Vanishing Gradients — Mastering ResNet and Skip ConnectionsHistorically, researchers assumed that continuously adding layers to a neural network would automatically yield better representations, as…
Madjid Korrabi·Jun 7Docling Part 4— EasyOCR Integration.In this blog, we’ll explore the pipeline of EasyOCR from the input image to the final text extraction.
Madjid Korrabi·May 19Docling Part 3— The AI Logic Behind Perfect Table Extraction.TableFormer in Docling: A Deep Dive into Architecture and Document Pipeline IntegrationA response icon1A response icon1
Madjid Korrabi·May 11Docling Part 2 — The Algorithmic Secrets Behind PDF Component Extraction.Introduction to RT-DETR Architecture
Madjid Korrabi·Apr 29CNN Part 1: Understanding Convolutional Neural Networks (CNNs) — From Image to PredictionHow does a computer look at an image and state with certainty: “That’s a zebra” and not a horse or a dog?