
Boosted K-Nearest Neighbor Search
Is it possible to integrate eDisMax-like boosting in Approximate Nearest Neighbor search for Solr and Lucene?

Is it possible to integrate eDisMax-like boosting in Approximate Nearest Neighbor search for Solr and Lucene?

This blogpost introduces Approximate Search Evaluator: a tool to measure vector search performance, addressing the accuracy/speed trade-off.

Embedding Model Evaluator is an MTEB benchmark designed to evaluate embedding models on user-provided datasets on retrieval and reranking tasks.

Dataset Generator automates the creation of relevance datasets for search evaluation, generating queries and relevance ratings with LLMs.

Vespa implements several useful features for customizing and improving Vector Search. Here, we will go into detail of each of them.

When using Lucene’s parent-child block-join mode, you are dealing with a “one-to-many relationship”. You have a single document (parent) owning multiple vector embeddings (children).To search

This blogpost explores the performance impact of DocValues vs. Inverted Index for Apache Solr facets done through JSON facet API.

Discover late interaction in Apache Solr: how to implement ColBERT-style neural reranking to boost search accuracy.

Discover late interaction in Apache Solr: how to implement ColBERT-style neural reranking to boost search accuracy.

This blog summarises the main new features introduced in Apache Solr 10.0.0, focusing on Vector Search and Learning to Rank (LTR).
We are Sease, an Information Retrieval Company based in London, focused on providing R&D project guidance and implementation, Search consulting services, Training, and Search solutions using open source software like Apache Lucene/Solr, Elasticsearch, OpenSearch and Vespa.
WhatsApp us