cs.LGMay 3, 2026

Retrieval with Multiple Query Vectors through Anomalous Pattern Detection

Authors: Allassan Tchangmena A NkenBaimam Boukar Jean JacquesMiriam RateikeCelia CintasSkyler Speakman

Organizations: 1IBM · University of Galway · 3Carnegie Mellon University Africa · University of Tübingen

Abstract

A classical vector retrieval problem typically considers a \emph{single} query embedding vector as input and retrieves the most similar embedding vectors from a vector database. However, complex reasoning and retrieval tasks frequently require \emph{multiple query vectors}, rather than a single one. In this work, we propose a retrieval method that considers multiple query vectors simultaneously and retrieves the most relevant vectors from the database using concepts from anomalous pattern detection. Specifically, our approach leverages a set of query vectors QQ (with Q1|Q|\geq 1), and identifies the subset of vector dimensions within QQ that standout (anomalous) from the rest of dimensions. Next, we scan the vector database to retrieve the set of vectors that are also anomalous across the previously identified vector dimensions and return them as our retrieved set of vectors. We validate our approach on two image datasets, a text dataset, and a tabular dataset. Overall, we observe that, across most datasets, larger query sets lead to improved retrieval performance. The improvement is most pronounced when increasing the query sets from 1 to 8, while the gains become smaller beyond that.

Explore similar work

CardsList
  1. QASP: Query-Adaptive Robust Vector Search Policy

    Jul 31, 2026Hakan Ferhatosmanoglu, Kushal Kumar, Tal Wagner +1Vector DatabaseSearch Algorithms

  2. Semantic Recall for Vector Search

    Apr 22, 2026Leonardo Kuffo, Ioanna Tsakalidou, Roberta De Viti +3Approximate Nearest-Neighbor SearchRetrieval Recall