GAS: A Lightweight Framework for Filtered Search over Wide-table Vectors
Summary: GAS enables filtered ANN over wide-table vectors using one attribute-agnostic graph plus lightweight, query-log-derived adaptive shortcuts. It avoids per-attribute index costs while preserving traversal efficiency, delivering up to 42.1× speedups with thousands of attributes. (summarized by gpt-5.6-luna on Aug 28 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. Ziyuan He (Beihang University)
- 2. Yuxiang Wang (Beihang University)
- 3. Yu Sun (Nankai University)
- 4. Zijie Ma (Beihang University)
- 5. Hui Li (Xidian University)
- 6. Qian Tao (Beihang University)
- 7. Yu Li (Beihang University)
- 8. Yongxin Tong (Beihang University)
BibTeX Citation
@article{he_vldb26,
title = {{GAS: A Lightweight Framework for Filtered Search over Wide-table Vectors}},
author = {He, Ziyuan and Wang, Yuxiang and Sun, Yu and Ma, Zijie and Li, Hui and Tao, Qian and Li, Yu and Tong, Yongxin},
journal = {PVLDB},
series = {{VLDB} '26},
volume = {19},
number = {11},
pages = {3133--3145},
doi = {10.14778/3836663.3836678},
url = {https://doi.org/10.14778/3836663.3836678},
year = {2026}
}
Incoming Citations (Sorted by Pagerank)
Showing 0 of 0 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 25 of 25 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
Previous
Page 1 / 1
Next