Improving Matrix-vector Multiplication via Lossless Grammar-Compressed Matrices
Summary: Proposes a lossless matrix compressor for real-valued data with mat-vec time/space proportional to compressed size, beating gzip and approaching xz. Adds column-reordering via a similarity score to cut peak memory and beat CLA in multithreaded speed. (summarized by gpt-5-nano on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Paolo Ferragina (University of Pisa)
- 2. Giovanni Manzini (University of Pisa)
- 3. Travis Gagie (Dalhousie University)
- 4. Dominik Köppl (Tokyo Medical and Dental University)
- 5. Gonzalo Navarro (University of Chile)
- 6. Manuel Striani (University of Eastern Piedmont)
- 7. Francesco Tosoni (University of Pisa)
BibTeX Citation
@article{ferragina_vldb22,
title = {{Improving Matrix-vector Multiplication via Lossless Grammar-Compressed Matrices}},
author = {Ferragina, Paolo and Manzini, Giovanni and Gagie, Travis and Köppl, Dominik and Navarro, Gonzalo and Striani, Manuel and Tosoni, Francesco},
journal = {PVLDB},
series = {{VLDB} '22},
volume = {15},
number = {10},
pages = {2175--2187},
doi = {10.14778/3547305.3547321},
url = {https://doi.org/10.14778/3547305.3547321},
year = {2022}
}Incoming Citations (Sorted by Pagerank)
Showing 3 of 3 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 8,794 | AWARE: Workload-aware, Redundancy-exploiting Linear Algebra | 2023 | SIGMOD | 5.370464e-05 |
| 10,589 | Morphing-based Compression for Data-centric ML Pipelines | 2026 | VLDB | 5.093636e-05 |
| 10,666 | HyperMR: Efficient Hypergraph-enhanced Matrix Storage on Compute-in-Memory Architecture | 2025 | SIGMOD | 5.093636e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 3 of 3 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 1,644 | Compressed Linear Algebra for Large-Scale Machine Learning | 2016 | VLDB | 0.00010132912 |
| 3,912 | Compressing Large Boolean Matrices Using Reordering Techniques | 2004 | VLDB | 7.0231849e-05 |
| 5,985 | Column Partition and Permutation for Run Length Encoding in Columnar Databases | 2020 | SIGMOD | 6.0185035e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 925 | Dictionary-based Order-preserving String Compression for Main Memory Column Stores | 2009 | SIGMOD |
| 2 | 10,589 | Morphing-based Compression for Data-centric ML Pipelines | 2026 | VLDB |
| 3 | 8,247 | Optimal Approximate Matrix Multiplication over Sliding Windows | 2026 | VLDB |
| 4 | 9,766 | On Efficient Large Sparse Matrix Chain Multiplication | 2024 | SIGMOD |
| 5 | 3,912 | Compressing Large Boolean Matrices Using Reordering Techniques | 2004 | VLDB |
| 6 | 9,520 | Experimental Analysis of Large-scale Learnable Vector Storage Compression | 2024 | VLDB |
| 7 | 6,485 | Tuple-oriented Compression for Large-scale Mini-batch Stochastic Gradient Descent | 2019 | SIGMOD |
| 8 | 8,794 | AWARE: Workload-aware, Redundancy-exploiting Linear Algebra | 2023 | SIGMOD |
| 9 | 6,046 | Automatic Optimization of Matrix Implementations for Distributed Machine Learning and Linear Algebra | 2021 | SIGMOD |
| 10 | 1,644 | Compressed Linear Algebra for Large-Scale Machine Learning | 2016 | VLDB |