analiza aiqbench
This research introduces Semantic Pyramid Indexing (SPI), a novel vector database framework that dynamically adjusts retrieval depth based on query complexity to reduce latency in streaming RAG pipelines without requiring global index rebuilds. By organizing embeddings into multiple resolution levels and utilizing an uncertainty-aware controller, SPI achieves significant speedups over fixed-traversal baselines while maintaining competitive recall rates across standard benchmarks like MS MARCO. The approach offers a practical solution for scaling approximate nearest neighbor search efficiently as data ingestion and query processing occur concurrently in production environments.
SPI: Query-Depth-Adaptive Indexing for Streaming RAG in Vector Databases
arXiv:2511.16681v3 Announce Type: replace Abstract: Vector databases (VecDBs) are increasingly deployed in retrieval-augmented generation (RAG) pipelines where query processing and document ingestion occur concurrently. The index layer needs to provide low-latency search while incorporating new vectors without frequent global rebuilding. Existing VecDB pipelines typically operate within a uniform representation regime, despite substantial variation in the semantic granularity required across queries. This motivates an index design that supports incremental updates while adapting retrieval depth to query distribution and complexity. We propose \textbf{Semantic Pyramid Indexing (SPI)}, a VecDB-layer indexing framework that organizes embeddings into $L$ semantically aligned resolution levels and selects retrieval depth per query via a lightweight uncertainty-aware controller. SPI supports progressive coarse-to-fine ANN search, level-wise streaming insertion without global rebuilds, and distributed execution through LSH partitioning with asynchronous gRPC coordination. Unlike hierarchical ANN structures with fixed traversal rules (e.g., SPANN), SPI adapts resolution at query time while remaining compatible with FAISS and Qdrant backends. On MS MARCO and Natural Questions, SPI achieves competitive Recall@10 with lower latency under the same dense encoder family, yielding a \textbf{1.4--2.3$\times$} average retrieval latency reduction under fixed Recall@10 targets relative to comparable approximate-ANN baselines. A prototype scaling study up to 8 nodes shows $6.2\times$ throughput scaling (${\approx}73\%$ efficiency); the 16-node configuration is included for completeness but shows diminishing efficiency. We provide a top-$K$ stability guarantee: queries with sufficient retrieval margin return an identical top-$K$ set at a shallower level. Code and configurations are available at https://github.com/FastLM/SPI_VecDB.
Abstract:Vector databases (VecDBs) are increasingly deployed in retrieval-augmented generation (RAG) pipelines where query processing and document ingestion occur concurrently. The index layer needs to provide low-latency search while incorporating new vectors without frequent global rebuilding. Existing VecDB pipelines typically operate within a uniform representation regime, despite substantial variation in the semantic granularity required across queries. This motivates an index design that supports incremental updates while adapting retrieval depth to query distribution and complexity. We propose \textbf{Semantic Pyramid Indexing (SPI)}, a VecDB-layer indexing framework that organizes embeddings into $L$ semantically aligned resolution levels and selects retrieval depth per query via a lightweight uncertainty-aware controller. SPI supports progressive coarse-to-fine ANN search, level-wise streaming insertion without global rebuilds, and distributed execution through LSH partitioning with asynchronous gRPC coordination. Unlike hierarchical ANN structures with fixed traversal rules (e.g., SPANN), SPI adapts resolution at query time while remaining compatible with FAISS and Qdrant backends. On MS MARCO and Natural Questions, SPI achieves competitive Recall@10 with lower latency under the same dense encoder family, yielding a \textbf{1.4--2.3$\times$} average retrieval latency reduction under fixed Recall@10 targets relative to comparable approximate-ANN baselines. A prototype scaling study up to 8 nodes shows $6.2\times$ throughput scaling (${\approx}73\%$ efficiency); the 16-node configuration is included for completeness but shows diminishing efficiency. We provide a top-$K$ stability guarantee: queries with sufficient retrieval margin return an identical top-$K$ set at a shallower level. Code and configurations are available at this https URL.
Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
arXiv:2511.16681 [cs.CL]
(or arXiv:2511.16681v3 [cs.CL] for this version)
Text integral sub licența deschisă a publicației · 253 cuvinte
Subiecte conexe
KnowledgeForge: mining gold from the ITSM ticket graveyard
AWS · 19 aug. 2026
The payload: the semantic layer, memory, and the loop that fills it
Neo4j · 10 aug. 2026
Building ambient agents with Amazon Bedrock AgentCore: From event-driven signals to human-in-the-loop workflows
AWS · 1 oct. 2026
Selecting a vector store for Amazon Bedrock Knowledge Bases
AWS · 17 sept. 2026