Graph-based Approximate Nearest Neighbor (ANN) indexes, such as HNSW, are the backbone of modern vector search. However, their massive memory footprint creates a severe infrastructure bottleneck. While vector quantization is widely adopted to reduce the memory required to store data vectors, the graph's edge structure...
Francesco Benocci, Leonard Dahlmann, Francesco Gargiulo et al.· Proceedings of the 35th ACM...· 1 citation
This paper investigates how to substantially reduce the memory footprint of learned sparse retrieval indexes without sacrificing the efficiency of state-of-the-art retrieval data structures. Building on SEISMIC, we revisit both levels of its design: the inverted index used to select candidates and the forward index use...
Franco Maria Nardini, Luca Rizzo, Cosimo Rulli et al.· 0 citations
FastPair is presented, a GPU decoder that optimizes the existing dictionary decoding process by reorganizing lookups and assembling decoded substrings for contiguous output writes, reaching up to 1.6 TB/s on a B300.
J. Isaacs, Francesco Gargiulo, P. Boncz et al.· arXiv.org· 0 citations
PUMA, a sparse autoencoder recipe that maps universal multimodal embeddings to compact sparse codes without retraining the backbone, reduces vector storage by 8-16x (FP32) and is up to 25x faster than exact dense scoring on larger candidate pools, enabling efficient multimodal retrieval.
Matteo Attimonelli, Alessandro De Bellis, Franco Maria Nardini et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
