Information Retrieval

Inverted indexes and compression.

1. Postings Lists as Sorted DocID Arrays
2. Term Dictionary Lookup Structures for Inverted Indexes
3. Delta Encoding and Variable-Byte Postings
4. Bit Packing and Frame-of-Reference Blocks
5. Document-at-a-Time and Term-at-a-Time Evaluation
6. Front Coding the Term Dictionary

Scoring models and feature design.

1. BM25 Term Saturation and Document Length Normalization
2. Disjunction-Max Combination for Multi-Field Scoring
3. Score Normalization and Rank Fusion Across Retrievers
4. Static Rank Priors and Their Effect on Ranking
5. Query-Document Feature Extraction for Learned Ranking
6. Click Features and Position Bias in Ranked Lists

Tokenization, expansion, and intent.

1. Unicode Word Boundary Rules in Query Tokenizers
2. Token Positions and Offsets in Analyzer Output
3. Stop-Word Removal and the Phrase Queries It Breaks
4. Suffix-Stripping Stemmers and the Conflations They Force
5. Synonym Expansion at Index Time Versus Query Time
6. Edit-Distance Candidate Generation for Query Spelling Correction

Relevance judgments and ranking metrics.

1. How Relevance Judgments Are Encoded in a qrels File
2. Precision and Recall at a Cutoff: Why the Cutoff Changes the Metric
3. Average Precision and the Mean Over Topics
4. Graded Gains and the NDCG Discount
5. Reciprocal Rank and Rank-Biased Precision
6. Pooling Depth and the Unjudged Tail

Embeddings and approximate nearest neighbors.

1. Cosine Similarity and the Geometry of Normalized Embedding Vectors
2. Distance Concentration and Why Tree Indexes Collapse in High Dimensions
3. Random Hyperplane LSH and the Arithmetic of Banding
4. Coarse Quantization with Inverted Files and the nprobe Setting
5. Product Quantization and the Asymmetric Distance Computation
6. HNSW Layer Assignment and the Greedy Descent

Scheduling, politeness, and freshness.

1. The URL Frontier: Priority Queues Behind Per-Host Politeness
2. Matching Paths in robots.txt: Longest Rule, Not First Rule
3. Per-Host Request Rate Limits and the Cost of a Fixed Delay
4. URL Normalization as a Set-Equality Problem
5. Conditional GET and the Revalidation Economy
6. Estimating Change Rates from Interval-Censored Fetches

Index updates and query latency.

1. The Document Add Path into an Inverted Index
2. Segment-Based Indexing and the Merge Operation
3. Deletion via Tombstones and the Delete Bitvector
4. Near-Real-Time Refresh and the Commit Boundary
5. Query-Time Cost and the Conjunctive Fast Path
6. Query and Filter Caches in a Search Engine

Result presentation and interaction.

1. Generating Search Result Snippets with Query-Biased Summaries
2. Relevance Feedback Through Query Expansion and Term Reweighting
3. Faceted Navigation and the Counts on Each Filter
4. Spelling Correction and the Did-You-Mean Suggestion
5. Query Autocomplete and the Prefix Trie Behind It
6. Pagination, Ranking Stability, and the Shuffling Result Set