Query Transformation Techniques: Improving RAG Recall
Query transformation is a high-leverage improvement for RAG systems.
23 articles
Query transformation is a high-leverage improvement for RAG systems.
Cross-encoder reranking typically improves RAG precision by 10-20%.
Hybrid search continues to outperform pure vector or keyword approaches. Invest in tuning your fusion strategy for your specific domain.
Vector databases in 2025 are more capable, efficient, and integrated than ever. Choose based on your specific requirements for scale, latency, and…
Hybrid search delivers better results than either approach alone. Implement it early in your RAG pipeline and tune the weights based on your specific use case.
Expanded Vector Dimensions Semantic Ranking Improvements Customer-Managed Keys for Vectors
Taking Vector Search to production requires careful consideration of performance, reliability, and maintenance. This guide covers production-ready patterns.
Databricks Vector Search enables semantic similarity search over your lakehouse data. Build RAG applications, recommendation systems, and intelligent search…
While context precision measures noise in retrieved results, context recall measures completeness. Are you retrieving all the documents needed to fully…
Context precision measures whether the retrieved documents are actually relevant to answering the question. High precision means less noise for the…
The retrieval component of RAG systems directly impacts generation quality. This guide provides a comprehensive overview of retrieval metrics and how to…
Retrieval-Augmented Generation (RAG) systems combine retrieval and generation components, each requiring specific evaluation strategies. This guide covers…
Cohere's models are now available on Azure AI, offering specialized capabilities for enterprise search and retrieval-augmented generation (RAG). This guide…
Production search indexes require operational discipline. From index versioning to blue-green deploys, these are the ops patterns I use to avoid downtime…
Bing Chat represents the future of search: not just finding links, but synthesizing answers from the web with AI understanding.
Cross encoders directly score query document pairs: Use an LLM to judge relevance: Using Cohere's specialized rerank API: 1. Retrieve more, re rank fewer :…
A robust way to combine rankings: Adjust weights based on query characteristics: 1. Start balanced : 50/50 is often a good default 2. Tune on your data :…
Azure Cognitive Search provides enterprise-grade search capabilities with AI enrichment options.
Azure SQL to Cognitive Search is a combination I've set up half a dozen times, and the reason it keeps appearing is that most enterprise organisations…
Custom skills are Azure Functions that implement a specific contract.
Default Cognitive Search will get you to "decent enough." Excellent search is a tuning exercise, and the levers that matter are mostly hidden. Custom…
Indexing PDFs is easy. Indexing PDFs in a way that makes them findable is a different sport. Cognitive Search skillsets are the part I usually sell to…
Azure Cognitive Search provides powerful search capabilities that can transform how users find and discover content in your applications.