DBScholar

Back to papers

CGIF: Combining Proximity Graphs and Inverted Files for Efficient Filtered Vector Search over Arbitrary Predicates

Summary: CGIF augments HNSW with predicate-aware local exploration, replacing filtered-out neighbors using inverted-file candidates while preserving graph navigation. It supports arbitrary predicates and selectivities, achieving up to 2× faster filtered ANNS with high recall. (summarized by gpt-5.6-luna on Aug 28 2026)

Paper ID
he8ed3e86007eebf2
Venue
VLDB
Year
2026
Pagerank
4.9793485e-05
Overall Rank
10,888 | 26.80%
DOI
10.14778/3836663.3836716

Incoming Non-self Citations Over Time

No non-self incoming citations found for this paper in this database.

Authors

BibTeX Citation

@article{luo_vldb26,
        title = {{CGIF: Combining Proximity Graphs and Inverted Files for Efficient Filtered Vector Search over Arbitrary Predicates}},
        author = {Luo, Jiarui and Zuo, Chaoji and Deng, Dong},
        journal = {PVLDB},
        series = {{VLDB} '26},
        volume = {19},
        number = {11},
        pages = {3663--3675},
        doi = {10.14778/3836663.3836716},
        url = {https://doi.org/10.14778/3836663.3836716},
        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 8 of 8 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Previous Page 1 / 1 Next

Semantically Similar Papers