Retrieval, Knowledge, and Memory

Discover why LLMs need Retrieval-Augmented Generation. Learn how RAG bridges knowledge gaps, reduces hallucinations, and enables non-parametric memory.

Master RAG system design by exploring retriever-generator interactions, timing strategies like iterative retrieval, and architectural variations like RETRO.

Master dense retrieval for semantic search. Explore bi-encoder architectures, embedding metrics, and contrastive learning to overcome keyword limitations.

Master contrastive learning for dense retrieval. Learn to train models using InfoNCE loss, in-batch negatives, and hard negative mining strategies effectively.

Master document chunking for RAG systems. Explore fixed-size, recursive, and semantic strategies to balance retrieval precision with context window limits.

Learn how embedding models convert text to vectors for RAG. Covers bi-encoder architecture, pooling strategies, dimensionality trade-offs, and model selection.

Explore vector similarity search for RAG systems. Compare cosine, dot product, and Euclidean metrics, and implement exact vs. approximate search with FAISS.

Master Hierarchical Navigable Small World (HNSW) graphs for vector search. Learn graph architecture, construction, and tuning for high-speed retrieval.

Master IVF indexes for scalable vector search. Learn clustering-based partitioning, nprobe tuning, and IVF-PQ compression for billion-scale retrieval.

Learn how Product Quantization compresses embeddings up to 100x using learned codebooks and asymmetric distance computation for scalable vector search.

Learn how hybrid search fuses BM25 keyword retrieval with dense vector retrieval using reciprocal rank fusion and weighted score combination to improve recall.

Learn how reranking with cross-encoders solves bi-encoder limitations. Master two-stage retrieval, training strategies, and latency optimization for production search systems.

Master RAG prompt engineering with strategic context placement, citation formats, and truncation strategies to improve LLM accuracy and reduce hallucinations.

Master RAG evaluation with metrics for retrieval quality (Precision@K, NDCG, MRR) and generation faithfulness using the RAGAS framework for AI systems.

What models store, what retrieval stores, and how to choose among prompting, RAG, editing, and retraining.

Localized weight updates, memory-based editors, specificity, generalization, and multi-hop consistency.

Time-sensitive facts, temporal benchmarks, update propagation, versioning, and rollback.

User models, episodic and semantic memory, consent, retention, conflict resolution, and forgetting.

Forget sets, retraining baselines, approximate removal, privacy goals, and the limits of verification.

Efficacy, locality, generalization, side effects, privacy leakage, and auditable change histories.

Covers continual learning definition, catastrophic forgetting, continual learning scenarios.

Covers elastic weight consolidation, synaptic intelligence, parameter importance, regularization trade-offs.

Covers replay buffer design, pseudo-rehearsal, generative replay, replay selection.

Covers progressive networks, expert expansion, architecture search, modular approaches.

Covers forward transfer, backward transfer, evaluation protocols, continual benchmarks.