Back to papers
Can Large Language Models Be Query Optimizer for Relational Databases?
Summary: Explores using LLMs as query optimizers by autoregressively generating PostgreSQL execution plans from serialized DB metadata, queries and plans (QInstruct), avoiding explicit plan enumeration. Proposes two-stage fine-tuning (Qit + Qdpo) and shows LLM-QO yields valid, high-quality plans that outperform traditional and learned optimizers on three workloads, suggesting strong generalization and adaptivity.
(summarized by gpt-5-mini on Feb 11 2026)
- Paper ID
- 7368
- Venue
- SIGMOD
- Year
- 2026
- Pagerank
- 4.4967983e-05
- Overall Rank
- 8,479 | 41.08%
- DOI
-
10.1145/3769771
Incoming Non-self Citations Over Time
Incoming Citations (Sorted by Pagerank)
Showing 3 of 3 citing papers.
Outgoing Citations (Sorted by Pagerank)
Showing 19 of 19 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank |
Cited Paper |
Year |
Venue |
Pagerank |
| 1 |
Access Path Selection in a Relational Database Management System |
1979 |
SIGMOD |
0.0040465394 |
| 71 |
How Good Are Query Optimizers, Really? |
2016 |
VLDB |
0.00059446482 |
| 329 |
Neo: A Learned Query Optimizer |
2019 |
VLDB |
0.00027301488 |
| 454 |
An Overview of Query Optimization in Relational Systems |
1998 |
PODS |
0.00022796106 |
| 634 |
Bao: Making Learned Query Optimization Practical |
2021 |
SIGMOD |
0.00018844568 |
| 997 |
CAESURA: Language Models as Multi-Modal Query Planners |
2024 |
CIDR |
0.00014726927 |
| 1,088 |
Language Models Enable Simple Systems for Generating Structured Views of Heterogeneous Data Lakes |
2024 |
VLDB |
0.00014158762 |
| 2,090 |
Balsa: Learning a Query Optimizer Without Expert Demonstrations |
2022 |
SIGMOD |
9.5668285e-05 |
| 2,585 |
Table-GPT: Table Fine-tuned GPT for Diverse Table Tasks |
2024 |
SIGMOD |
8.4909917e-05 |
| 2,937 |
DSB: A Decision Support Benchmark for Workload-Driven and Traditional Database Systems |
2021 |
VLDB |
7.8552033e-05 |
| 3,105 |
GPTuner: A Manual-Reading Database Tuning System via GPT-Guided Bayesian Optimization |
2024 |
VLDB |
7.5567226e-05 |
| 3,236 |
TPC-DS, Taking Decision Support Benchmarking to the Next Level |
2002 |
SIGMOD |
7.3294564e-05 |
| 3,345 |
Lero: A Learning-to-Rank Query Optimizer |
2023 |
VLDB |
7.1908499e-05 |
| 3,465 |
LLM-R2: A Large Language Model Enhanced Rule-based Rewrite System for Boosting Query Efficiency |
2025 |
VLDB |
7.0668293e-05 |
| 3,729 |
Cost-based or Learning-based? A Hybrid Query Optimizer for Query Plan Selection |
2022 |
VLDB |
6.8078013e-05 |
| 5,924 |
Data-Juicer: A One-Stop Data Processing System for Large Language Models |
2024 |
SIGMOD |
5.2674873e-05 |
| 6,712 |
Demonstrating GPT-DB: Generating Query-Specific and Customizable Code for SQL Processing with GPT-4 |
2023 |
VLDB |
4.9474017e-05 |
| 7,009 |
Is Your Learned Query Optimizer Behaving As You Expect? A Machine Learning Perspective |
2024 |
VLDB |
4.8597992e-05 |
| 8,157 |
Automated Data Visualization from Natural Language via Large Language Models: An Exploratory Study |
2024 |
SIGMOD |
4.5701714e-05 |
Semantically Similar Papers
| Overall Rank |
Paper |
Year |
Venue |
Pagerank |
| 8,660 |
Learned Offline Query Planning via Bayesian Optimization |
2025 |
SIGMOD |
4.4680058e-05 |
| 10,219 |
Practical Parameterized Query Optimization via Efficient Plan Reuse and List-wise Ranking |
2026 |
SIGMOD |
4.1905499e-05 |
| 11,352 |
DeepO: A Learned Query Optimizer |
2022 |
SIGMOD |
4.1905499e-05 |
| 10,112 |
SEFRQO: A Self-Evolving Fine-Tuned RAG-Based Query Optimizer |
2026 |
SIGMOD |
4.1905499e-05 |
| 3,465 |
LLM-R2: A Large Language Model Enhanced Rule-based Rewrite System for Boosting Query Efficiency |
2025 |
VLDB |
7.0668293e-05 |
| 10,884 |
RankPQO: Learning-to-Rank for Parametric Query Optimization |
2025 |
VLDB |
4.1905499e-05 |
| 10,217 |
This is Going to Sound Crazy, But What If We Used Large Language Models to Boost Automatic Database Tuning Algorithms By Leveraging Prior History? We Will Find Better Configurations More Quickly Than Retraining From Scratch! |
2026 |
SIGMOD |
4.1905499e-05 |
| 5,429 |
Logical and Physical Optimizations for SQL Query Execution over Large Language Models |
2025 |
SIGMOD |
5.511638e-05 |
| 9,992 |
Leveraging Query Optimizers to Verify the Soundness of LLM-based Query Rewrites for Real-World Workloads, and More! |
2026 |
CIDR |
4.1905499e-05 |
| 7,009 |
Is Your Learned Query Optimizer Behaving As You Expect? A Machine Learning Perspective |
2024 |
VLDB |
4.8597992e-05 |