Concepts

The IR fundamentals that underpin everything in search. Don’t try to read all of these at once — for a guided foundation order, see How to Start a Career in Search. If you learn only three: BM25, Precision and Recall, and how to measure a ranking change (NDCG + Judgment Lists).

A

• A-B Testing for Search · ACORN-1 · APD · ASH · Approximate Nearest Neighbor Search · Agentic Query Workload · Agentic Search · Asymmetric Semantic Search · Autocomplete

B

• BBQ · BERT · BM25 · Bag-of-Documents Model · Bayesian BM25 · Bi-Encoder · Binary Quantization · Block-Max WAND · Brute-Force Vector Search

C

• Clean Context · Click Models · Click Signals · Clicks Residual · ColBERT · ColPali · Collocations · Compute-Storage Disaggregation · Consistency Filtering · Context Engineering · Contrastive Gap · Contrastive Learning · Conversational Search · Cross-Encoder

D

• Dense Embeddings · Dense Passage Retriever · Dense Vector Retrieval · Dimensionality Reduction · Direct Corpus Interaction · Diversity Metrics

E

• ELSER · Economics of Search · Embedding Fine-tuning · Embeddings

F

• FLAN-T5 · Faceted Search · Feature Store · Full-Text Search

G

• GGUF

H

• HNSW · Hard Negative Mining · Hashing Trick · Hit Rate at K · Hybrid Search · Hypothetical Document Embeddings

I

• ITQ · IVF · Implicit Judgments · Impression Bias · Intent Drift · Interleaving · Isolated Feedback Loops

J

• Judgment Lists

K

• Keywords Extraction · Knowledge Distillation · Knowledge Graph Search

L

• LSH · LLM as Judge · LTR Feature Engineering · LambdaLoss · LambdaMART · Late Interaction · Learned Sparse Retrieval · Learning to Rank · Linear Score Combination · Listwise Relevance Evaluation · LoRA

M

• MAP · MMR · MRR · MUVERA · Matryoshka Embeddings · Modality Gap · MonoT5 · Multimodal Embeddings · miniCOIL

N

• NDCG
• Neural Click Models

O

• Out-of-Time Validation · Out-of-Vocabulary

P

• PCA · PEFT · PROMPTAGATOR · Pairwise Relevance Evaluation · Personalization · Pointwise Relevance Evaluation · Pooling · Position Bias · Precision and Recall · Presentation Bias · Purpose-Built Agentic Search Models

Q

• QLoRA · Query Expansion · Query Relaxation · Query Sampling · Query Segmentation · Query Specificity · Query Types · Query Understanding

R

• RAG · RaBitQ · RankGPT · RankLLaMA · RankNet · Ranking Objectives · Ranking Signal Selection · Reciprocal Rank Fusion · Reinforcement Learning for Search · Relative Score Fusion · Relevance Feedback · Reranking · Results Boosting · Results Merchandising · Retrieval Pipeline

S

• SIRA · SPLADE · Scalar Quantization · Score Normalization · Search Architecture · Search Evaluation · Search Governance · Search Intent · Search Results Explainability · Search Scopes · Search Team · Search-R1 · Semantic Boosting · Semantic Search · Session-Based Evaluation · Sharding · Signal Downboosting · Sparse Embeddings · Sparse Vector Retrieval · Spelling Correction · Staged Judging · Statistical Significance in Search Evaluation · Stopwords · Synonyms · Synthetic Query Generation

T

• Task-Aware Embeddings · Text Chunking · Token Pooling · Tokenization · TurboQuant · t-SNE

U

• UDCG · UMAP

V

• Vector Filtering · Vector Quantization · Vector Similarity Metrics

W

• WAND · Wormhole Vectors

Z

• Zero Results · Zero-Shot Retrieval

Related

• Topics · Tools · Case Studies · People