Efficient Dense Vector Search within Knowledge Graph Content Embeddings

2026-09-21

AI Summary

"This study addresses the challenge of efficient dense vector search within RDF knowledge graphs to support multimodal queries and learning-based similarity ranking. By extending the QLever system, the work introduces native support for dense vector retrieval, incorporating three key optimizations: engine-level tensor function registration, vocabulary-time parsing of JSON-encoded vectors, and exposure of vector index via a virtual SERVICE. Additionally, two novel benchmarks were developed. The implementation, named QUIVER, facilitates cross-modal vector joins and demonstrates significant performance improvements. Specifically, vocabulary-time parsing alone achieves speedups of up to 41.9x (BSBM) and 20x (DBpedia), with further enhancements to 355x (BSBM) and 97.8x (DBpedia) when approximate nearest neighbor indexing is utilized."

Abstract

Knowledge graphs are a core component of today's knowledge infrastructure, supporting reasoning and anchoring knowledge systems to verifiable facts. RDF stores and SPARQL engines fulfill this function, enabling a range of retrieval and inference tasks on structured knowledge. Coupling them with Language Models (LMs) extends RAG toward neurosymbolic reasoning, where structured queries gate or re-rank generative outputs. This line of reasoning requires that SPARQL evaluation natively support tensor operations on dense embeddings, enabling multimodal querying and learned similarity-based ranking to be expressed together with graph-structural constraints. This approach is feasible only if the engine can efficiently perform dense vector search. We present QLever-Unified Indexed Vector Embedding Retrieval (QUIVER), an extension to QLever that adds native support for dense vector retrieval within RDF knowledge graphs. It implements three optimizations: engine-level registration of tensor functions, vocabulary-time parsing of JSON-encoded vectors, and a virtual SERVICE that exposes a vector index inside the query. We propose two new benchmarks: an extension of Berlin Sparql Benchmark (BSBM) with text embeddings and an extension of DBpedia with image embeddings. Against the baselines, vocabulary-time parsing alone yields median speedups of up to 41.9x on BSBM and 20x on DBpedia for single-type ranking; adding an approximate nearest-neighbor index yields speedups of 355x on BSBM and 97.8x on DBpedia. The index further makes cross-modal vector joins on DBpedia feasible in seconds, whereas all non-indexed configurations time out.

Problem

• Dense Vector Search
• Knowledge Graphs
• Neurosymbolic Reasoning
• Tensor Operations
• SPARQL

Innovation

• Dense Vector Search
• Knowledge Graphs
• Tensor Operations
• Approximate Nearest-Neighbor Index
• Multimodal Querying