Proves that for bounded n, there exist MV embeddings with m vectors each whose Chamfer matrix requires single-vector dimension D = (ε² m)^Ω(1/ε) to approximate within ε, separating MV from SV expressiveness.
arXiv preprint arXiv:2405.19504 , year =
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
abstract
Neural embedding models have become a fundamental component of modern information retrieval (IR) pipelines. These models produce a single embedding $x \in \mathbb{R}^d$ per data-point, allowing for fast retrieval via highly optimized maximum inner product search (MIPS) algorithms. Recently, beginning with the landmark ColBERT paper, multi-vector models, which produce a set of embedding per data point, have achieved markedly superior performance for IR tasks. Unfortunately, using these models for IR is computationally expensive due to the increased complexity of multi-vector retrieval and scoring. In this paper, we introduce MUVERA (MUlti-VEctor Retrieval Algorithm), a retrieval mechanism which reduces multi-vector similarity search to single-vector similarity search. This enables the usage of off-the-shelf MIPS solvers for multi-vector retrieval. MUVERA asymmetrically generates Fixed Dimensional Encodings (FDEs) of queries and documents, which are vectors whose inner product approximates multi-vector similarity. We prove that FDEs give high-quality $\epsilon$-approximations, thus providing the first single-vector proxy for multi-vector similarity with theoretical guarantees. Empirically, we find that FDEs achieve the same recall as prior state-of-the-art heuristics while retrieving 2-5$\times$ fewer candidates. Compared to prior state of the art implementations, MUVERA achieves consistently good end-to-end recall and latency across a diverse set of the BEIR retrieval datasets, achieving an average of 10$\%$ improved recall with $90\%$ lower latency.
years
2026 5representative citing papers
ANN search quality is better assessed by 1/Ratio@k than Recall@k because the former tracks downstream task utility more closely while allowing substantially lower computational cost.
CAMI frames multi-index construction for semantic retrieval as a budgeted multi-objective portfolio problem and uses agent-guided search plus confidence-aware pruning to find high-recall configurations with reduced evaluation cost.
Diagnosable ColBERT aligns ColBERT embeddings to an expert-grounded clinical latent space to enable direct diagnosis of model misunderstandings and better training data curation.
ClinicalAligner26AM tops the MultiClinCorpus shared task by distilling Sinkhorn-sharpened multi-level alignments into a clinical encoder for projecting Spanish entity annotations to six target languages with F1 above 0.95.
citing papers explorer
-
Multi-Vector Embeddings are Provably More Expressive than Single Vector Embeddings
Proves that for bounded n, there exist MV embeddings with m vectors each whose Chamfer matrix requires single-vector dimension D = (ε² m)^Ω(1/ε) to approximate within ε, separating MV from SV expressiveness.
-
ANN Search: Recall What Matters
ANN search quality is better assessed by 1/Ratio@k than Recall@k because the former tracks downstream task utility more closely while allowing substantially lower computational cost.
-
CAMI: Cost-Aware Agent-Guided Multi-Indexing for Semantic Retrieval
CAMI frames multi-index construction for semantic retrieval as a budgeted multi-objective portfolio problem and uses agent-guided search plus confidence-aware pruning to find high-recall configurations with reduced evaluation cost.
-
Diagnosable ColBERT: Debugging Late-Interaction Retrieval Models Using a Learned Latent Space as Reference
Diagnosable ColBERT aligns ColBERT embeddings to an expert-grounded clinical latent space to enable direct diagnosis of model misunderstandings and better training data curation.
-
ClinicalAligner26AM: A Cross-Lingual Aligner for Dataset Translation; Evidences from the MultiClinCorpus Shared Task
ClinicalAligner26AM tops the MultiClinCorpus shared task by distilling Sinkhorn-sharpened multi-level alignments into a clinical encoder for projecting Spanish entity annotations to six target languages with F1 above 0.95.