GleanVec: Accelerating vector search with minimalist nonlinear dimensionality reduction
This addresses the challenge of compute and memory pressure in vector search systems for applications such as cross-modal retrieval, though it appears incremental as it builds on existing dimensionality reduction techniques.
The paper tackles the problem of accelerating high-dimensional vector search for applications like cross-modal retrieval by introducing new linear and nonlinear dimensionality reduction methods, LeanVec-Sphering and GleanVec, which improve search accuracy and computational efficiency, with initial results showing they push the state of the art.
Embedding models can generate high-dimensional vectors whose similarity reflects semantic affinities. Thus, accurately and timely retrieving those vectors in a large collection that are similar to a given query has become a critical component of a wide range of applications. In particular, cross-modal retrieval (e.g., where a text query is used to find images) is gaining momentum rapidly. Here, it is challenging to achieve high accuracy as the queries often have different statistical distributions than the database vectors. Moreover, the high vector dimensionality puts these search systems under compute and memory pressure, leading to subpar performance. In this work, we present new linear and nonlinear methods for dimensionality reduction to accelerate high-dimensional vector search while maintaining accuracy in settings with in-distribution (ID) and out-of-distribution (OOD) queries. The linear LeanVec-Sphering outperforms other linear methods, trains faster, comes with no hyperparameters, and allows to set the target dimensionality more flexibly. The nonlinear Generalized LeanVec (GleanVec) uses a piecewise linear scheme to further improve the search accuracy while remaining computationally nimble. Initial experimental results show that LeanVec-Sphering and GleanVec push the state of the art for vector search.