From HNSW To Information-theoretic Binarization: Rethinking The Architecture Of Scalable Vector Search

Abstract

Modern semantic search and retrieval-augmented generation (RAG) systems rely predominantly on in-memory approximate nearest neighbor (ANN) indexes over high-precision floating-point vectors, resulting in escalating operational cost and inherent trade-offs between latency, throughput, and retrieval accuracy. This paper analyzes the architectural limitations of the dominant "HNSW + float32 + cosine similarity" stack and evaluates existing cost-reduction strategies, including storage disaggregation and lossy vector quantization, which inevitably sacrifice either performance or accuracy. We introduce and empirically evaluate an alternative information-theoretic architecture based on maximally informative binarization (MIB), efficient bitwise distance metrics, and an information-theoretic scoring (ITS) mechanism. Unlike conventional ANN systems, this approach enables exhaustive search over compact binary representations, allowing deterministic retrieval and eliminating accuracy degradation

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyabtahi2025from

Related papers

• Passing The Baton: High Throughput Distributed Disk-based Vector Search With Batann (2025)0.00
• Aisaq: All-in-storage ANNS With Product Quantization For Dram-free Information Retrieval (2024)0.00
• Vectorsearch: Enhancing Document Retrieval With Semantic Embeddings And Optimized Search (2024)0.00
• Lossless Compression Of Vector Ids For Approximate Nearest Neighbor Search (2025)11.11
• Practical And Asymptotically Optimal Quantization Of High-dimensional Vectors In Euclidean Space For Approximate Nearest Neighbor Search (2024)8.82
• Efficient And Effective Retrieval Of Dense-sparse Hybrid Vectors Using Graph-based Approximate Nearest Neighbor Search (2024)0.00
• Operational Advice For Dense And Sparse Retrievers: HNSW, Flat, Or Inverted Indexes? (2024)0.00
• Ultra-high Dimensional Sparse Representations With Binarization For Efficient Text Retrieval (2021)8.60