SEFRQO: A Self-Evolving Fine-Tuned RAG-Based Query Optimizer

Summary: SEFRQO is a self-evolving RAG-based query optimizer that fine-tunes LLMs (supervised + reinforcement) to produce performance-oriented query hints by retrieving execution-feedback. It dynamically builds prompts from similar queries and per-query execution records for continual in-context learning, cutting latency up to ~65–94% vs prior LQOs. (summarized by gpt-5-mini on Feb 11 2026)

Paper ID: hddeec3dcd78afb60
Venue: SIGMOD
Year: 2026
Pagerank: 4.9769913e-05
Overall Rank: 10,607 | 28.71%
DOI: 10.1145/3769826
PDF: Download (CC BY 4.0)

Incoming Non-self Citations Over Time

No non-self incoming citations found for this paper in this database.

Authors

1. Hanwen Liu (University of Southern California)
2. Qihan Zhang (University of Southern California)
3. Ryan Marcus (University of Pennsylvania)
4. Ibrahim Sabek (University of Southern California)

BibTeX Citation

@inproceedings{liu_sigmod26,
        title = {{SEFRQO: A Self-Evolving Fine-Tuned RAG-Based Query Optimizer}},
        author = {Liu, Hanwen and Zhang, Qihan and Marcus, Ryan and Sabek, Ibrahim},
        series = {{SIGMOD} '26},
        booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
        publisher = {Association for Computing Machinery},
        doi = {10.1145/3769826},
        url = {https://dl.acm.org/doi/10.1145/3769826},
        year = {2026}
}

Incoming Citations (Sorted by Pagerank)

Showing 2 of 2 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
10,892	QDBO: A Real-time Quantum-augmented Database System Optimizer	2026	VLDB	4.9769913e-05
10,950	Ultron: History-Based Query Optimization at Databricks	2026	VLDB	4.9769913e-05

Outgoing Citations (Sorted by Pagerank)

Showing 22 of 22 cited papers.

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

Rank	Cited Paper	Year	Venue	Pagerank
15	How Good Are Query Optimizers, Really?	2016	VLDB	0.00061067652
144	Neo: A Learned Query Optimizer	2019	VLDB	0.00029090793
174	Text-to-SQL Empowered by Large Language Models: A Benchmark Evaluation	2024	VLDB	0.00026787121
189	Milvus: A Purpose-Built Vector Data Management System	2021	SIGMOD	0.0002585319
361	Bao: Making Learned Query Optimization Practical	2021	SIGMOD	0.00020000855
655	CAESURA: Language Models as Multi-Modal Query Planners	2024	CIDR	0.00015090153
1,195	Balsa: Learning a Query Optimizer Without Expert Demonstrations	2022	SIGMOD	0.00011574218
1,245	DB-BERT: A Database Tuning Tool that "Reads the Manual"	2022	SIGMOD	0.0001136308
1,735	Flow-Loss: Learning Cardinality Estimates That Matter	2021	VLDB	9.7566604e-05
2,209	Lero: A Learning-to-Rank Query Optimizer	2023	VLDB	8.8360101e-05
2,227	GPTuner: A Manual-Reading Database Tuning System via GPT-Guided Bayesian Optimization	2024	VLDB	8.8007923e-05
2,686	Cost-based or Learning-based? A Hybrid Query Optimizer for Query Plan Selection	2022	VLDB	8.1300913e-05
3,479	LOGER: A Learned Optimizer towards Generating Efficient and Robust Query Execution Plans	2023	VLDB	7.2665349e-05
3,948	Deploying a Steered Query Optimizer in Production at Microsoft	2022	SIGMOD	6.9051584e-05
4,191	Kepler: Robust Learning for Faster Parametric Query Optimization	2023	SIGMOD	6.7425275e-05
4,240	LEON: A New Framework for ML-Aided Query Optimization	2023	VLDB	6.7064546e-05
4,384	R-Bot: An LLM-based Query Rewrite System	2025	VLDB	6.6232918e-05
5,236	FASTgres: Making Learned Query Optimizer Hinting Effective	2023	VLDB	6.2153504e-05
5,776	Lemo: A Cache-Enhanced Learned Optimizer for Concurrent Queries	2023	SIGMOD	5.9957692e-05
6,072	Demonstrating GPT-DB: Generating Query-Specific and Customizable Code for SQL Processing with GPT-4	2023	VLDB	5.8936535e-05
6,575	Can Large Language Models Be Query Optimizer for Relational Databases?	2026	SIGMOD	5.7428777e-05
8,021	Generating Succinct Descriptions of Database Schemata for Cost-Efficient Prompting of Large Language Models	2024	VLDB	5.4041713e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	3,717	Logical and Physical Optimizations for SQL Query Execution over Large Language Models	2025	SIGMOD
2	10,899	RAGPerf: An End-to-End Benchmarking Framework for Retrieval-Augmented Generation Systems	2026	VLDB
3	6,298	Is Your Learned Query Optimizer Behaving As You Expect? A Machine Learning Perspective	2024	VLDB
4	2,037	LLM-R^2: A Large Language Model Enhanced Rule-based Rewrite System for Boosting Query Efficiency	2025	VLDB
5	10,697	SchemaRAG: A Schema-aware Retrieval-Augmented Generation Framework for Text-to-SQL	2026	SIGMOD
6	4,384	R-Bot: An LLM-based Query Rewrite System	2025	VLDB
7	10,371	Leveraging Query Optimizers to Verify the Soundness of LLM-based Query Rewrites for Real-World Workloads, and More!	2026	CIDR
8	10,790	SafeQL: Search-based Refinement for Safe and Efficient LLM-based Text-to-SQL	2026	VLDB
9	10,842	ReSequel: Robust LLM-assisted Query Rewriting and Optimization using Templatization and Sampling	2026	VLDB
10	6,575	Can Large Language Models Be Query Optimizer for Relational Databases?	2026	SIGMOD