QStore: Quantization-Aware Compressed Model Storage
Summary: QStore losslessly co-stores low- and high-precision foundation models as the low-precision model plus compact residuals, avoiding redundant files. Lightweight decoding preserves fast access, delivering up to 2.2× storage savings and 1.8× faster loading. (summarized by gpt-5.6-luna on Jul 24 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Raunak Shah (University of Illinois Urbana-Champaign)
- 2. Zhaoheng Li (University of Illinois Urbana-Champaign)
- 3. Yongjoo Park (University of Illinois Urbana-Champaign)
BibTeX Citation
@article{shah_vldb26,
title = {{QStore: Quantization-Aware Compressed Model Storage}},
author = {Shah, Raunak and Li, Zhaoheng and Park, Yongjoo},
journal = {PVLDB},
series = {{VLDB} '26},
volume = {19},
number = {3},
pages = {388--398},
doi = {10.14778/3778092.3778100},
url = {https://doi.org/10.14778/3778092.3778100},
year = {2026}
}Incoming Citations (Sorted by Pagerank)
Showing 2 of 2 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 10,898 | stratum: A System Infrastructure for Massive Agent-Centric ML Workloads | 2026 | VLDB | 4.9793485e-05 |
| 11,047 | Chipmink: Efficient Delta Identification for Massive Object Graphs | 2026 | VLDB | 4.9793485e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 10 of 10 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 654 | Materialization Optimizations for Feature Selection Workloads | 2014 | SIGMOD | 0.0001510357 |
| 1,614 | Compressed Linear Algebra for Large-Scale Machine Learning | 2016 | VLDB | 0.00010071891 |
| 2,928 | ALP: Adaptive Lossless floating-Point Compression | 2023 | SIGMOD | 7.8424335e-05 |
| 3,140 | DeepSqueeze: Deep Semantic Compression for Tabular Data | 2020 | SIGMOD | 7.6023603e-05 |
| 6,592 | Tuple-oriented Compression for Large-scale Mini-batch Stochastic Gradient Descent | 2019 | SIGMOD | 5.7421684e-05 |
| 7,564 | SIEVE: Effective Filtered Vector Search with Collection of Indexes | 2025 | VLDB | 5.4955249e-05 |
| 8,532 | AirIndex: Versatile Index Tuning Through Data and Storage | 2023 | SIGMOD | 5.3217474e-05 |
| 10,151 | Kishu: Time-Traveling for Computational Notebooks | 2025 | VLDB | 5.0715586e-05 |
| 10,279 | ElasticNotebook: Enabling Live Migration for Computational Notebooks | 2024 | VLDB | 5.0455234e-05 |
| 11,147 | Demo of Kishu: Time-Traveling for Computational Notebooks | 2025 | SIGMOD | 4.9793485e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 13,576 | Waiting to Decompress: The Economics of LLM-Based Compression | 2026 | CIDR |
| 2 | 6,311 | Progressive Compressed Records: Taking a Byte out of Deep Learning Data | 2021 | VLDB |
| 3 | 9,700 | Experimental Analysis of Large-scale Learnable Vector Storage Compression | 2024 | VLDB |
| 4 | 10,776 | Efficient Cooperation-Aware Key and Value Management for LLM Inference | 2026 | VLDB |
| 5 | 11,274 | QPET: A Versatile and Portable Quantity-of-Interest-Preservation Framework for Error-Bounded Lossy Compression | 2025 | VLDB |
| 6 | 9,987 | Frequency-Store: Scaling Image AI by A Column-Store for Images | 2025 | CIDR |
| 7 | 9,071 | Privacy and Accuracy-Aware AI/ML Model Deduplication | 2025 | SIGMOD |
| 8 | 4,498 | PQCache: Product Quantization-based KVCache for Long Context LLM Inference | 2025 | SIGMOD |
| 9 | 8,791 | Everything You Always Wanted to Know About Storage Compressibility of Pre-Trained ML Models but Were Afraid to Ask | 2024 | VLDB |
| 10 | 10,582 | NeurStore: Efficient In-database Deep Learning Model Management System | 2026 | SIGMOD |