# AI-Driven Research for Databases

arXiv: https://arxiv.org/abs/2604.06566 · HTML: https://mlsanity.com/paper/2604.06566/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 79 references that are in the map.
- First posted on arXiv 8 Apr 2026.

## Builds on (most cited first, 26 of 26)

- [Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agents](https://mlsanity.com/paper/2505.22954/) (2025)
- [The Case for Learned Index Structures](https://mlsanity.com/paper/1712.01208/) (2017)
- [ShinkaEvolve: Towards Open-Ended And Sample-Efficient Program Evolution](https://mlsanity.com/paper/2509.19349/) (2025)
- [MLGym: A New Framework and Benchmark for Advancing AI Research Agents](https://mlsanity.com/paper/2502.14499/) (2025)
- [Neo: A Learned Query Optimizer](https://mlsanity.com/paper/1904.03711/) (2019)
- [Learned Cardinalities: Estimating Correlated Joins with Deep Learning](https://mlsanity.com/paper/1809.00677/) (2018)
- [ThetaEvolve: Test-time Learning on Open Problems](https://mlsanity.com/paper/2511.23473/) (2025)
- [Bao: Learning to Steer Query Optimizers](https://mlsanity.com/paper/2004.03814/) (2020)
- [ALEX: An Updatable Adaptive Learned Index](https://mlsanity.com/paper/1905.08898/) (2019)
- [An End-to-End Learning-based Cost Estimator](https://mlsanity.com/paper/1906.02560/) (2019)
- [Deep Unsupervised Cardinality Estimation](https://mlsanity.com/paper/1905.04278/) (2019)
- [Self-Improving Language Models for Evolutionary Program Synthesis: A Case Study on ARC-AGI](https://mlsanity.com/paper/2507.14172/) (2025)
- [GPTuner: A Manual-Reading Database Tuning System via GPT-Guided Bayesian Optimization](https://mlsanity.com/paper/2311.03157/) (2023)
- [Lero: A Learning-to-Rank Query Optimizer](https://mlsanity.com/paper/2302.06873/) (2023)
- [Apache Calcite: A Foundational Framework for Optimized Query Processing Over Heterogeneous Data Sources](https://mlsanity.com/paper/1802.10233/) (2018)
- [Code Researcher: Deep Research Agent for Large Systems Code and Commit History](https://mlsanity.com/paper/2506.11060/) (2025)
- [DB-BERT: a Database Tuning Tool that "Reads the Manual"](https://mlsanity.com/paper/2112.10925/) (2021)
- [Towards Dynamic and Safe Configuration Tuning for Cloud Databases](https://mlsanity.com/paper/2203.14473/) (2022)
- [LLM-R2: A Large Language Model Enhanced Rule-based Rewrite System for Boosting Query Efficiency](https://mlsanity.com/paper/2404.12872/) (2024)
- [R-Bot: An LLM-based Query Rewrite System](https://mlsanity.com/paper/2412.01661/) (2024)
- [Glia: A Human-Inspired AI for Automated Systems Design and Optimization](https://mlsanity.com/paper/2510.27176/) (2025)
- [LlamaTune: Sample-Efficient DBMS Configuration Tuning](https://mlsanity.com/paper/2203.05128/) (2022)
- [Updatable Learned Index with Precise Positions](https://mlsanity.com/paper/2104.05520/) (2021)
- [LOCAT: Low-Overhead Online Configuration Auto-Tuning of Spark SQL Applications](https://mlsanity.com/paper/2203.14889/) (2022)
- [Polyjuice: High-Performance Transactions via Learned Concurrency Control](https://mlsanity.com/paper/2105.10329/) (2021)
- [Accelerating String-Key Learned Index Structures via Memoization-based Incremental Training](https://mlsanity.com/paper/2403.11472/) (2024)

## Built on by (most cited first, 2 of 2)

- [AIChilles: Automatically Uncovering Hidden Weaknesses in AI-Evolved Systems](https://mlsanity.com/paper/2606.15834/) (2026)
- [Assistants, Not Architects: The Role of LLMs in Networked Systems Design](https://mlsanity.com/paper/2604.25506/) (2026)

## Abstract (arXiv)

As the complexity of modern workloads and hardware increasingly outpaces human research and engineering capacity, existing methods for database performance optimization struggle to keep pace. To address this gap, a new class of techniques, termed AI-Driven Research for Systems (ADRS), uses large language models to automate solution discovery. This approach shifts optimization from manual system design to automated code generation. The key obstacle, however, in applying ADRS is the evaluation pipeline. Since these frameworks rapidly generate hundreds of candidates without human supervision, they depend on fast and accurate feedback from evaluators to converge on effective solutions. Building such evaluators is especially difficult for complex database systems. To enable the practical application of ADRS in this domain, we propose automating the design of evaluators by co-evolving them with the solutions. We demonstrate the effectiveness of this approach through three case studies optimizing buffer management, query rewriting, and index selection. Our automated evaluators enable the discovery of novel algorithms that outperform state-of-the-art baselines (e.g., a deterministic query rewrite policy that achieves up to 6.8x lower latency), demonstrating that addressing the evaluation bottleneck unlocks the potential of ADRS to generate highly optimized, deployable code for next-generation data systems.