SIEVE: Effective Filtered Vector Search with Collection of Indexes

Summary: SIEVE replaces predicate-constrained graph traversal with a workload-aware collection of specialized indexes for filtered vector search. A three-dimensional size–latency–recall model guides index selection, yielding up to 8.06× speedups with modest memory overhead. (summarized by gpt-5.6-luna on Jul 24 2026)

Paper ID: hdae29a5ad781eaf3
Venue: VLDB
Year: 2025
Pagerank: 5.4929234e-05
Overall Rank: 7,570 | 49.13%
DOI: 10.14778/3749646.3749725
PDF: Download (CC BY-NC-ND 4.0)

Incoming Non-self Citations Over Time

Authors

1. Zhaoheng Li (University of Illinois Urbana-Champaign)
2. Silu Huang (ByteDance)
3. Wei Ding (ByteDance)
4. Yongjoo Park (University of Illinois Urbana-Champaign)
5. Jianjun Chen (ByteDance)

BibTeX Citation

@article{li_vldb25,
        title = {{SIEVE: Effective Filtered Vector Search with Collection of Indexes}},
        author = {Li, Zhaoheng and Huang, Silu and Ding, Wei and Park, Yongjoo and Chen, Jianjun},
        journal = {PVLDB},
        series = {{VLDB} '25},
        volume = {18},
        number = {11},
        pages = {4723--4736},
        doi = {10.14778/3749646.3749725},
        url = {https://doi.org/10.14778/3749646.3749725},
        year = {2025}
}

Incoming Citations (Sorted by Pagerank)

Showing 6 of 6 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
10,150	QStore: Quantization-Aware Compressed Model Storage	2026	VLDB	5.0691578e-05
10,421	An In-Depth Study of Filter-Agnostic Vector Search on a PostgreSQL Database System: [Experiments & Analysis]	2026	SIGMOD	4.9769913e-05
10,468	FAVOR: Efficient Filter-Agnostic Vector ANNS Based on Selectivity-Aware Exclusion Distances	2026	SIGMOD	4.9769913e-05
10,860	GAS: A Lightweight Framework for Filtered Search over Wide-table Vectors	2026	VLDB	4.9769913e-05
10,938	SQL-Native Vector Search at Billion Scale in Presto	2026	VLDB	4.9769913e-05
11,056	Chipmink: Efficient Delta Identification for Massive Object Graphs	2026	VLDB	4.9769913e-05

Outgoing Citations (Sorted by Pagerank)

Showing 21 of 21 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Rank	Cited Paper	Year	Venue	Pagerank
74	Fast Approximate Nearest Neighbor Search With The Navigating Spreading-out Graph	2019	VLDB	0.00037145866
88	Automated Selection of Materialized Views and Indexes for SQL Databases	2000	VLDB	0.00035340164
189	Milvus: A Purpose-Built Vector Data Management System	2021	SIGMOD	0.0002585319
252	Database Cracking	2007	CIDR	0.00023101361
339	AnalyticDB-V: A Hybrid Analytical Engine Towards Query Fusion for Structured and Unstructured Data	2020	VLDB	0.00020561716
379	Apache Calcite: A Foundational Framework for Optimized Query Processing Over Heterogeneous Data Sources	2018	SIGMOD	0.00019507406
553	Optimizing Queries Using Materialized Views: A Practical, Scalable Solution	2001	SIGMOD	0.00016518678
923	Materialized View Selection and Maintenance Using Multi-Query Optimization	2001	SIGMOD	0.00013059066
1,036	Fine-grained Partitioning for Aggressive Data Skipping	2014	SIGMOD	0.00012372946
1,128	Qd-tree: Learning Data Layouts for Big Data Analytics	2020	SIGMOD	0.00011901941
1,225	ACORN: Performant and Predicate-Agnostic Search Over Vector Embeddings and Structured Data	2024	SIGMOD	0.00011447937
1,373	High-Throughput Vector Similarity Search in Knowledge Graphs	2023	SIGMOD	0.00010891169
1,600	High-Dimensional Approximate Nearest Neighbor Search: with Reliable and Efficient Distance Comparison Operations	2023	SIGMOD	0.00010111099
1,747	Selecting Subexpressions to Materialize at Datacenter Scale	2018	VLDB	9.7303647e-05
1,958	SeRF: Segment Graph for Range-Filtering Approximate Nearest Neighbor Search	2024	SIGMOD	9.3118048e-05
2,086	An Architecture for Recycling Intermediates in a Column-store	2009	SIGMOD	9.0677901e-05
3,083	Skipping-oriented Partitioning for Columnar Layouts	2017	VLDB	7.6620866e-05
4,177	GraphJet: Real-Time Content Recommendations at Twitter	2016	VLDB	6.7540362e-05
4,600	Steiner-Hardness: A Query Hardness Measure for Graph-Based ANN Indexes	2024	VLDB	6.5055567e-05
8,568	Materialized View Selection for XQuery Workloads	2012	SIGMOD	5.3114964e-05
9,999	SparkCruise: Handsfree Computation Reuse in Spark	2019	VLDB	5.0964027e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	8,039	VecFlow: A High-Performance Vector Data Management System for Filtered-Search on GPUs	2026	SIGMOD
2	1,373	High-Throughput Vector Similarity Search in Knowledge Graphs	2023	SIGMOD
3	2,872	Graph-Based Vector Search: An Experimental Evaluation of the State-of-the-Art	2025	SIGMOD
4	10,639	Curator: Efficient Vector Search with Low-Selectivity Filters	2026	SIGMOD
5	10,455	Efficient Index Layout and Search Strategy for Large-scale High-dimensional Vector Similarity Search	2026	SIGMOD
6	8,117	Efficient and Tunable Similar Set Retrieval	2001	SIGMOD
7	2,061	Similarity search in the blink of an eye with compressed indices	2023	VLDB
8	10,648	Efficient Vector Index Merging in Vector Databases	2026	SIGMOD
9	1,958	SeRF: Segment Graph for Range-Filtering Approximate Nearest Neighbor Search	2024	SIGMOD
10	10,897	CGIF: Combining Proximity Graphs and Inverted Files for Efficient Filtered Vector Search over Arbitrary Predicates	2026	VLDB