NeuralPath LogoNeuralPath

NLP

Explore how machines tokenize, embed, and model human language from word vectors to sequence models.

Your Progress0 / 5 completed
0%
Continue Learning

Foundations

(1 topic)

Text Preprocessing and Tokenization

How machines translate human language into numbers—from classical text cleaning to subword Byte-Pair Encoding (BPE) that powers GPT-4, Claude, and Llama-3.

Language Modeling

(1 topic)

N-Grams and Language Models

How statistical language modeling began—from the Chain Rule of Probability and Markov N-Grams to Laplace Smoothing, Perplexity, and the leap to Neural LLMs.

Representations

(1 topic)

Word Embeddings: How AI Turns Words into Meaning

Learn how Word Embeddings turn words into coordinates of meaning—from simple personality sliders and Word2Vec to Cosine Similarity, PyTorch nn.Embedding, and modern LLM embeddings.

Sequence Modeling

(1 topic)

Sequence Models: RNNs, LSTMs, and GRUs Explained Simply

Learn how AI reads sentences word-by-word with memory—from simple Recurrent Neural Networks (RNNs) and the game-of-telephone problem to LSTMs, GRUs, and PyTorch code.

Evaluation & Benchmarks

(1 topic)

Evaluation for Language Tasks: Perplexity, BLEU, ROUGE & LLM-as-a-Judge

Learn how we grade AI on human language—from Perplexity, BLEU, and ROUGE to BERTScore, LLM-as-a-Judge, and blind human arenas.