Tagged articles
BM25
31 articles · Page 1 of 1
This article benchmarks keyword (BM25), vector (two Chinese embedding models), and hybrid retrieval on FastAPI's Chinese documentation, finding keyword search excels for exact-name queries while a larger vector model catches paraphrased questions, and hybrid helps only with strong embeddings; vector databases are unnecessary until data exceeds in-memory capacity.
BM25Embedding ModelsFastAPI
0 likes · 21 min read

The article analyzes context-mode, an MCP plugin that reduces AI coding agent context usage by 96% across 21 real-world scenarios by intercepting tool outputs, executing analysis in sandboxes, and indexing content in SQLite FTS5 with BM25 retrieval for session continuity.
AI programmingBM25Context Management
0 likes · 12 min read


Tech BeanSep 22, 2026 · Artificial Intelligence
This article provides a comprehensive technical analysis of Retrieval-Augmented Generation (RAG), covering embedding principles, chunking strategies, vector databases, dense/sparse/hybrid retrieval, reranking, HyDE, GraphRAG, and engineering evaluation frameworks with code examples.
BM25GraphRAGHybrid Search
0 likes · 24 min read


DabaoshiSep 21, 2026 · Artificial Intelligence
This article explains how to improve RAG retrieval quality by combining vector and keyword search via RRF fusion, adding cross-encoder reranking, rewriting user queries for better retrieval, and using HyDE to generate hypothetical answers for embedding-based search, with a practical Java implementation example.
BM25Cross-EncoderHyDE
0 likes · 19 min read

This guide walks through building a production-ready hybrid search system using Spring Boot and OpenSearch, covering mapping design for BM25 and k-NN vector search, synonym configuration, RRF-based score fusion, connection pool tuning, multi-tenant security with DLS/FLS, CDC data sync, and cluster operations including ISM lifecycle policies.
BM25CDCHybrid Search
0 likes · 18 min read

The article examines advanced data‑warehouse RAG challenges, explains why pure vector retrieval falls short, details a BM25‑plus‑vector hybrid approach, showcases AI‑driven metric platforms and data‑agent design, and outlines practical factors for building a robust data‑agent foundation.
AI agentsBM25LLM
0 likes · 7 min read

Hybrid search combines dense vector retrieval with sparse keyword search, using RRF fusion and optional reranking, to overcome the limitations of each method—semantic understanding versus exact matching—making it the production‑grade standard for RAG systems by 2025‑2026.
BM25ElasticsearchHybrid Search
0 likes · 19 min read

QMD is an open‑source, MIT‑licensed local search engine written in TypeScript that combines BM25, vector embeddings via a GGUF model, and an LLM reranker, allowing natural‑language queries over thousands of markdown files without network calls, and can be installed with just three commands.
BM25LLM rerankTypeScript
0 likes · 13 min read

The article evaluates four non‑vector Retrieval‑Augmented Generation methods—BM25 lexical search, GraphRAG graph traversal, Tree‑Search document navigation, and an Agentic search loop—using a small JSON‑based corpus, showing each method’s strengths, weaknesses, and when to combine them for production‑grade retrieval.
Agentic SearchBM25GraphRAG
0 likes · 12 min read

This tutorial explains Retrieval‑Augmented Generation (RAG) from fundamentals to a full pipeline, covering text chunking strategies, VoyageAI embeddings, vector‑store implementation, BM25 lexical search, and a multi‑index retriever that fuses semantic and lexical results with Reciprocal Rank Fusion.
BM25PythonRAG
0 likes · 48 min read

The article examines why pure vector retrieval in RAG lacks lexical precision and traceable relevance scores, explains BM25's strengths, and presents hybrid retrieval architectures—including RRF and linear combination fusion—as well as the trade‑offs of externalizing the fusion process.
BM25Hybrid SearchRAG
0 likes · 9 min read

The article analyzes how BM25 excels at exact token matching while vector embeddings capture semantic intent, explains their distinct failure modes, and shows that a hybrid retriever—combined with metadata filtering, proper chunking, and reciprocal rank fusion—delivers the most reliable results for RAG pipelines.
BM25EmbeddingHybrid Retrieval
0 likes · 17 min read

This article breaks down why RAG pipelines often underperform and shows how proper chunking, overlap settings, hybrid vector‑plus‑BM25 retrieval, and a Rerank step can dramatically improve recall and precision, with concrete code examples and tuning tips.
BM25Hybrid RetrievalLangChain
0 likes · 14 min read

This article explains PostgreSQL's native full‑text search, its core concepts of tsvector and tsquery, demonstrates how to use built‑in functions and operators, compares built‑in, zhparser, and pg_search extensions for Chinese tokenization, and provides best‑practice tips for indexing, triggers, and performance optimization.
BM25Chinese tokenizationPostgreSQL
0 likes · 14 min read

The article explains how TF‑IDF and BM25 compute term importance, compares their strengths and weaknesses, and shows how these sparse retrieval methods integrate with dense retrieval techniques such as DPR, SPLADE, and ColBERT in Retrieval‑Augmented Generation systems, concluding with a hybrid retrieval decision matrix.
BM25Hybrid RetrievalRAG
0 likes · 14 min read

This article dissects the shortcomings of pure vector retrieval, explains how BM25 complements it, compares weighted‑sum and Reciprocal Rank Fusion (RRF) strategies, shows experimental results that identify optimal weight and k values, and provides practical engineering tips for deploying hybrid search in RAG systems.
BM25Hybrid RetrievalRAG Systems
0 likes · 24 min read

This article explains why combining BM25 with dense vector search using Reciprocal Rank Fusion (RRF) improves recall for both exact‑term and semantic queries in a financial‑insurance document corpus, details the underlying algorithms, parameter choices such as k=60, provides Python implementations, and shows measurable performance gains in production.
BM25FaissHybrid Retrieval
0 likes · 28 min read

This guide introduces PHPVector, a pure‑PHP vector database that combines HNSW‑based approximate nearest‑neighbor search with BM25 full‑text ranking, showing installation, document insertion, vector and text queries, hybrid ranking modes, configuration options, distance metrics, tuning tips, and persistence mechanisms.
AIBM25HNSW
0 likes · 10 min read

This article walks through building a practical Elasticsearch demo that lets users type a single query like “nearby clean coffee shop” and get results by combining dense‑vector semantic search, geo filtering, BM25, and a hybrid RRF‑style ranking, with both LLM‑based structuring and a fallback hash‑based embedding.
BM25FlaskHybrid Search
0 likes · 10 min read

This article outlines five practical engineering strategies—multi‑vector retrieval, manual splitting and labeling, scalar enhancement, context augmentation, and dense‑sparse vector integration—that together address common RAG retrieval bottlenecks and dramatically improve recall stability and answer quality.
BM25LLMRAG
0 likes · 17 min read

Claude Code introduces MCP Tool Search, a lazy‑loading mechanism that dynamically loads only needed tools, cutting token usage by over 67,000 tokens in large MCP setups, preventing context bloat, improving performance, and offering developers regex and BM25 search options with defer_loading support.
BM25Claude CodeContext Management
0 likes · 6 min read


ITPUBDec 29, 2025 · Databases
VectorChord‑BM25 is a PostgreSQL extension that adds native BM25 ranking and tokenization, delivering up to three‑fold query‑per‑second improvements over ElasticSearch while maintaining comparable relevance scores, and includes detailed installation, usage examples, and performance analysis.
BM25PostgreSQLRanking
0 likes · 17 min read

This article provides a comprehensive technical guide to Elasticsearch’s core retrieval models—BM25 and TF‑IDF—while detailing modern vector‑based search using dense_vector, kNN, L2 and cosine distances, and demonstrates how to combine keyword and semantic results through hybrid search and Reciprocal Rank Fusion (RRF) with practical configuration examples.
BM25ElasticsearchKNN
0 likes · 42 min read

This article describes a low‑cost, privacy‑preserving chatbot for logistics that combines data cleaning, large‑model‑based data augmentation, BM25 and vector retrieval, a DNN rerank model, and LLM‑driven answer rewriting to deliver accurate, compliant automated responses.
AIBM25Data Augmentation
0 likes · 11 min read

Anthropic’s Contextual Retrieval augments traditional RAG with contextual embeddings and BM25, reducing retrieval failure rates by 49% (up to 67% with reranking), improving accuracy across domains, and lowering latency and cost through Claude’s prompt‑caching feature.
AIBM25Contextual Retrieval
0 likes · 11 min read


JavaEdgeOct 2, 2024 · Artificial Intelligence
This article presents a contextual retrieval technique that combines contextual embeddings and contextual BM25 to reduce RAG miss rates by up to 67%, explains the underlying methods, implementation steps, cost considerations, experimental results, and practical deployment guidance.
AIBM25Contextual Retrieval
0 likes · 17 min read


政采云技术May 12, 2022 · Fundamentals
This article explains Lucene's query types, the step‑by‑step query execution flow—including entry, rewrite, weight creation, scoring, and result collection—while providing code examples and performance considerations to help developers troubleshoot and optimize search performance.
BM25ElasticsearchJava
0 likes · 15 min read

This article explains the architecture and key algorithms behind a specialized exam‑question search engine, covering query parsing, BM25‑based relevance scoring, text‑image hybrid retrieval, Learning‑to‑Rank models, and practical optimizations for long queries and large K‑12 datasets.
BM25Rankingelastic search
0 likes · 12 min read

The article surveys core search‑engine technology—document hierarchy, flat and vertical inverted indexes, query operators for building and merging score lists, and ranking models from Boolean and BM25 to language‑model approaches like Indri—providing a foundational overview of information retrieval.
BM25information retrievalinverted index
0 likes · 14 min read

The article explains how to systematically improve Elasticsearch search relevance by fine‑tuning Query DSL with filters, phrase matching, and boosts, incorporating static scoring via function_score, adjusting BM25 similarity parameters, and using diagnostics like _explain to iteratively achieve higher ranking quality.
BM25BoostElasticsearch
0 likes · 17 min read

The article explains how static source‑code scanning, binary analysis, and advanced code‑search technologies—including incremental indexing, deduplication, real‑time Sphinx indexing, and BM25 ranking—can be combined to detect and remediate product‑level vulnerabilities early, thereby significantly raising software quality and reducing risk.
BM25Sphinxcode search
0 likes · 13 min read
