LeaFi: Data Series Indexes on Steroids with Learned Filters

Summary: LeaFi uses learned filters to boost pruning in tree-based data-series indexes. Models predict tight node-wise distance lower bounds for pruning, with train-time index building and query-time calibration to meet per-query recall targets; up to 20x pruning, 32x search speed at 99% recall.— (summarized by gpt-5-nano on Feb 09 2026)

Paper ID: 7108
Venue: SIGMOD
Year: 2025
Pagerank: 5.2755515e-05
Overall Rank: 9,377 | 35.67%
DOI: 10.1145/3709701

Incoming Non-self Citations Over Time

Authors

1. Qitong Wang (Université Paris Cité)
2. Ioana Ileana (Université Paris Cité)
3. Themis Palpanas (French University Institute; Université Paris Cité)

BibTeX Citation

@inproceedings{wang_sigmod25,
        title = {{LeaFi: Data Series Indexes on Steroids with Learned Filters}},
        author = {Wang, Qitong and Ileana, Ioana and Palpanas, Themis},
        series = {{SIGMOD} '25},
        booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
        publisher = {Association for Computing Machinery},
        doi = {10.1145/3709701},
        url = {https://dl.acm.org/doi/10.1145/3709701},
        year = {2025}
}

Incoming Citations (Sorted by Pagerank)

Showing 4 of 4 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
7,145	Subspace Collision: An Efficient and Accurate Framework for High-dimensional Approximate Nearest Neighbor Search	2025	SIGMOD	5.6908957e-05
9,360	DARTH: Declarative Recall Through Early Termination for Approximate Nearest Neighbor Search	2026	SIGMOD	5.2819088e-05
10,297	TaCo: Data-adaptive and Query-aware Subspace Collision for High-dimensional Approximate Nearest Neighbor Search	2026	SIGMOD	5.093636e-05
10,430	Honeybee: Efficient Role-based Access Control for Vector Databases via Dynamic Partitioning	2026	SIGMOD	5.093636e-05

Outgoing Citations (Sorted by Pagerank)

Showing 30 of 30 cited papers.

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

Rank	Cited Paper	Year	Venue	Pagerank
41	Fast Subsequence Matching in Time-Series Databases	1994	SIGMOD	0.00046675394
43	The Case for Learned Index Structures	2018	SIGMOD	0.00046060254
190	Locally Adaptive Dimensionality Reduction for Indexing Large Time Series Databases	2001	SIGMOD	0.00026105472
332	Query-Aware Locality-Sensitive Hashing for Approximate Nearest Neighbor Search	2016	VLDB	0.00020920444
873	Learning Multi-dimensional Indexes	2020	SIGMOD	0.00013481915
898	Querying and Mining of Time Series Data: Experimental Comparison of Representations and Distance Measures	2008	VLDB	0.00013339042
926	Improving Approximate Nearest Neighbor Search through Learned Adaptive Early Termination	2020	SIGMOD	0.00013181732
1,084	A Data-adaptive and Dynamic Segmentation Index for Whole Matching on Time Series	2013	VLDB	0.00012256753
1,174	Tsunami: A Learned Multi-dimensional Index for Correlated Data and Skewed Workloads	2021	VLDB	0.00011817414
1,579	k-Shape: Efficient and Accurate Clustering of Time Series	2015	SIGMOD	0.00010305183
2,004	TSB-UAD: An End-to-End Benchmark Suite for Univariate Time-Series Anomaly Detection	2022	VLDB	9.3207067e-05
2,534	ELPIS: Graph-Based Similarity Search for Scalable Data Science	2023	VLDB	8.4561875e-05
2,543	Learned Cardinality Estimation: An In-depth Study	2022	SIGMOD	8.4445934e-05
2,734	Return of the Lernaean Hydra: Experimental Evaluation of Data Series Approximate Similarity Search	2020	VLDB	8.190416e-05
2,888	AI Meets Database: AI4DB and DB4AI	2021	SIGMOD	7.9941489e-05
2,991	FactorJoin: A New Cardinality Estimation Framework for Join Queries	2023	SIGMOD	7.8880723e-05
3,037	The Lernaean Hydra of Data Series Similarity Search: An Experimental Evaluation of the State of the Art	2019	VLDB	7.8275859e-05
3,240	Graph-Based Vector Search: An Experimental Evaluation of the State-of-the-Art	2025	SIGMOD	7.6071649e-05
3,486	Scalable, Variable-Length Similarity Search in Data Series: The ULISSE Approach	2018	VLDB	7.3696676e-05
3,516	LOGER: A Learned Optimizer towards Generating Efficient and Robust Query Execution Plans	2023	VLDB	7.3524442e-05
3,809	Auto-WLM: Machine Learning Enhanced Workload Management in Amazon Redshift	2023	SIGMOD	7.1074195e-05
4,212	Data Series Progressive Similarity Search with Probabilistic Quality Guarantees	2020	SIGMOD	6.8283344e-05
4,617	Learned Cardinality Estimation for Similarity Queries	2021	SIGMOD	6.604437e-05
4,665	Coconut: A Scalable Bottom-Up Approach for Building Data Series Indexes	2018	VLDB	6.5780693e-05
4,883	Indexing for Interactive Exploration of Big Data Series	2014	SIGMOD	6.4648124e-05
5,231	Hercules Against Data Series Similarity Search	2022	VLDB	6.3068064e-05
5,800	DET-LSH: A Locality-Sensitive Hashing Scheme with Dynamic Encoding Tree for Approximate Nearest Neighbor Search	2024	VLDB	6.0850924e-05
6,691	Dumpy: A Compact and Adaptive Index for Large Data Series Collections	2023	SIGMOD	5.8011086e-05
9,278	Odyssey: A Journey in the Land of Distributed Data Series Similarity Search	2023	VLDB	5.2936898e-05
9,394	iEDeaL: A Deep Learning Framework for Detecting Highly Imbalanced Interictal Epileptiform Discharges	2023	VLDB	5.2755515e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	10,262	LINE: A Learned Index with Group-Enhanced Leaves and Cache-Optimized Inner Tree	2026	SIGMOD
2	9,964	DIDS: Double Indices and Double Summarizations for Fast Similarity Search	2024	VLDB
3	3,047	Similarity Evaluation on Tree-structured Data	2005	SIGMOD
4	3,792	Learned Index: A Comprehensive Experimental Evaluation	2023	VLDB
5	847	Benchmarking Learned Indexes	2021	VLDB
6	6,832	A Scalable Index for Top-k Subtree Similarity Queries	2019	SIGMOD
7	6,906	Efficient Similarity Join and Search on Multi-Attribute Data	2015	SIGMOD
8	11,702	LES3: Learning-based Exact Set Similarity Search	2021	VLDB
9	4,300	DILI: A Distribution-Driven Learned Index	2023	VLDB
10	10,469	LM-Tree: A Hybrid Learned Index for Similarity Search in Metric Spaces	2026	SIGMOD