# Adversarial Query Synthesis via Bayesian Optimization

arXiv: https://arxiv.org/abs/2603.01570 · HTML: https://mlsanity.com/paper/2603.01570/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 24 references that are in the map.
- First posted on arXiv 2 Mar 2026.

## Builds on (most cited first, 12 of 12)

- [The Power of Scale for Parameter-Efficient Prompt Tuning](https://mlsanity.com/paper/2104.08691/) (2021)
- [The Case for Learned Index Structures](https://mlsanity.com/paper/1712.01208/) (2017)
- [Protein Design with Guided Discrete Diffusion](https://mlsanity.com/paper/2305.20009/) (2023)
- [Design-Bench: Benchmarks for Data-Driven Offline Model-Based Optimization](https://mlsanity.com/paper/2202.08450/) (2022)
- [Local Latent Space Bayesian Optimization over Structured Inputs](https://mlsanity.com/paper/2201.11872/) (2022)
- [XGrammar: Flexible and Efficient Structured Generation Engine for Large Language Models](https://mlsanity.com/paper/2411.15100/) (2024)
- [Bao: Learning to Steer Query Optimizers](https://mlsanity.com/paper/2004.03814/) (2020)
- [Benchmarking Learned Indexes](https://mlsanity.com/paper/2006.12804/) (2020)
- [Learned Offline Query Planning via Bayesian Optimization](https://mlsanity.com/paper/2502.05256/) (2025)
- [Yannakakis+: Practical Acyclic Query Evaluation with Theoretical Guarantees](https://mlsanity.com/paper/2504.03279/) (2025)
- [Redbench: A Benchmark Reflecting Real Workloads](https://mlsanity.com/paper/2506.12488/) (2025)
- [JOB-Complex: A Challenging Benchmark for Traditional & Learned Query Optimization](https://mlsanity.com/paper/2507.07471/) (2025)

## Abstract (arXiv)

Benchmark workloads are extremely important to the database management research community, especially as more machine learning components are integrated into database systems. Here, we propose a Bayesian optimization technique to automatically search for difficult benchmark queries, significantly reducing the amount of manual effort usually required. In preliminary experiments, we show that our approach can generate queries with more than double the optimization headroom compared to existing benchmarks.