Zero-Shot Cost Models for Out-of-the-box Learned Cost Prediction
Summary: Zero-shot cost models deliver out-of-the-box learned cost estimation that generalizes to unseen databases without training queries. A novel architecture and workload encoding enable transfer from pre-trained models, outperforming workload-driven baselines and supporting few-shot refinement on new databases. (summarized by gpt-5-nano on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Benjamin Hilprecht (Technical University of Darmstadt)
- 2. Carsten Binnig (German National Research Center for Information Technology; Technical University of Darmstadt)
BibTeX Citation
@article{hilprecht_vldb22,
title = {{Zero-Shot Cost Models for Out-of-the-box Learned Cost Prediction}},
author = {Hilprecht, Benjamin and Binnig, Carsten},
journal = {PVLDB},
series = {{VLDB} '22},
volume = {15},
number = {11},
pages = {2361--2374},
doi = {10.14778/3551793.3551799},
url = {https://doi.org/10.14778/3551793.3551799},
year = {2022}
}Incoming Citations (Sorted by Pagerank)
Showing 37 of 37 citing papers.
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 17 of 17 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 3,967 | Efficient Deep Learning Pipelines for Accurate Cost Estimations Over Large Scale Query Workload | 2021 | SIGMOD |
| 2 | 5,918 | On Efficient Approximate Queries over Machine Learning Models | 2023 | VLDB |
| 3 | 320 | DeepDB: Learn from Data, not from Queries! | 2020 | VLDB |
| 4 | 1,064 | Are We Ready For Learned Cardinality Estimation? | 2021 | VLDB |
| 5 | 11,484 | Learned Cost Models for Query Optimization: From Batch to Streaming Systems | 2025 | VLDB |
| 6 | 453 | An End-to-End Learning-based Cost Estimator | 2020 | VLDB |
| 7 | 4,641 | How Good are Learned Cost Models, Really? Insights from Query Optimization Tasks | 2025 | SIGMOD |
| 8 | 5,854 | Modeling Shifting Workloads for Learned Database Systems | 2024 | SIGMOD |
| 9 | 6,158 | Rethinking Learned Cost Models: Why Start from Scratch? | 2023 | SIGMOD |
| 10 | 4,549 | One Model to Rule them All: Towards Zero-Shot Learning for Databases | 2022 | CIDR |