SAQ: Pushing the Limits of Vector Quantization through Code Adjustment and Dimension Segmentation
Summary: SAQ segments PCA-projected vectors and DP-optimizes bit allocation to favor high-magnitude segments, lowering quantization error. A coordinate-descent code-adjustment encodes dimensions independently then refines, avoiding enumeration and yielding up to 80% error reduction and >80× encoding speedup over Extended RabitQ. (summarized by gpt-5-mini on Feb 11 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Hui Li (Chinese University of Hong Kong)
- 2. Shiyuan Deng (Huawei)
- 3. Xiao Yan (Wuhan University)
- 4. Xiangyu Zhi (Chinese University of Hong Kong)
- 5. James Cheng (Chinese University of Hong Kong)
BibTeX Citation
@inproceedings{li_sigmod26,
title = {{SAQ: Pushing the Limits of Vector Quantization through Code Adjustment and Dimension Segmentation}},
author = {Li, Hui and Deng, Shiyuan and Yan, Xiao and Zhi, Xiangyu and Cheng, James},
series = {{SIGMOD} '26},
booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
publisher = {Association for Computing Machinery},
doi = {10.1145/3769824},
url = {https://dl.acm.org/doi/10.1145/3769824},
year = {2026}
}Incoming Citations (Sorted by Pagerank)
Showing 2 of 2 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 10,520 | TaCo: Data-adaptive and Query-aware Subspace Collision for High-dimensional Approximate Nearest Neighbor Search | 2026 | SIGMOD | 4.9769913e-05 |
| 10,839 | RNSG: A Range-Aware Graph Index for Efficient Range-Filtered Approximate Nearest Neighbor Search | 2026 | VLDB | 4.9769913e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 15 of 15 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
Previous
Page 1 / 1
Next