FAVOR: Efficient Filter-Agnostic Vector ANNS Based on Selectivity-Aware Exclusion Distances
Summary: FAVOR unifies selectivity estimation with filter-agnostic HNSW ANNS for arbitrary hybrid predicates. Its exclusion distances reshape search geometry, while selectivity-driven routing chooses pre-filtered brute force or HNSW, yielding 1.3–5× higher QPS at 95% recall. (summarized by gpt-5.6-luna on Jul 26 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. Junjie Song (Huazhong University of Science and Technology)
- 2. Yu Liu (Huazhong University of Science and Technology; National University of Singapore)
- 3. Guoyu Hu (National University of Singapore)
- 4. Zhongle Xie (Zhejiang University)
- 5. Ming Yang (National University of Singapore; Wuhan Technical University)
- 6. Beng Chin Ooi (Zhejiang University)
- 7. Ke Zhou (Huazhong University of Science and Technology)
BibTeX Citation
@inproceedings{song_sigmod26,
title = {{FAVOR: Efficient Filter-Agnostic Vector ANNS Based on Selectivity-Aware Exclusion Distances}},
author = {Song, Junjie and Liu, Yu and Hu, Guoyu and Xie, Zhongle and Yang, Ming and Ooi, Beng Chin and Zhou, Ke},
series = {{SIGMOD} '26},
booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
publisher = {Association for Computing Machinery},
doi = {10.1145/3802060},
url = {https://dl.acm.org/doi/10.1145/3802060},
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 22 of 22 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
Previous
Page 1 / 1
Next