Semantic Vector Encoding And Similarity Search Using Fulltext Search Engines

Abstract

Vector representations and vector space modeling (VSM) play a central role in modern machine learning. We propose a novel approach to `vector similarity searching' over dense semantic representations of words and documents that can be deployed on top of traditional inverted-index-based fulltext engines, taking advantage of their robustness, stability, scalability and ubiquity. We show that this approach allows the indexing and querying of dense vectors in text domains. This opens up exciting avenues for major efficiency gains, along with simpler deployment, scaling and monitoring. The end result is a fast and scalable vector database with a tunable trade-off between vector search performance and quality, backed by a standard fulltext engine such as Elasticsearch. We empirically demonstrate its querying performance and quality by applying this solution to the task of semantic searching over a dense vector representation of the entire English Wikipedia.

Stats

• citations7
• S2 citations—
• github stars0
• HF likes0
• heat score6.77
• arxiv keyrygl2017semantic

Related papers

• Vectorsearch: Enhancing Document Retrieval With Semantic Embeddings And Optimized Search (2024)0.00
• Evaluating The Impact Of Word Embeddings On Similarity Scoring In Practical Information Retrieval (2026)0.00
• Leanvec: Searching Vectors Faster By Making Them Fit (2023)0.00
• Lucene For Approximate Nearest-neighbors Search On Arbitrary Dense Vectors (2019)0.00
• Neural Vector Spaces For Unsupervised Information Retrieval (2017)12.93
• Gleanvec: Accelerating Vector Search With Minimalist Nonlinear Dimensionality Reduction (2024)0.00
• Reveal Hidden Pitfalls And Navigate Next Generation Of Vector Similarity Search From Task-centric Views (2025)0.00
• Optimization Of Latent-space Compression Using Game-theoretic Techniques For Transformer-based Vector Search (2025)0.00