Drift-adapter: A Practical Approach To Near Zero-downtime Embedding Model Upgrades In Vector Databases

Abstract

Upgrading embedding models in production vector databases typically requires re-encoding the entire corpus and rebuilding the Approximate Nearest Neighbor (ANN) index, leading to significant operational disruption and computational cost. This paper presents Drift-Adapter, a lightweight, learnable transformation layer designed to bridge embedding spaces between model versions. By mapping new queries into the legacy embedding space, Drift-Adapter enables the continued use of the existing ANN index, effectively deferring full re-computation. We systematically evaluate three adapter parameterizations: Orthogonal Procrustes, Low-Rank Affine, and a compact Residual MLP, trained on a small sample of paired old and new embeddings. Experiments on MTEB text corpora and a CLIP image model upgrade (1M items) show that Drift-Adapter recovers 95-99% of the retrieval recall (Recall@10, MRR) of a full re-embedding, adding less than 10 microseconds of query latency. Compared to operational strategies l

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyvejendla2025drift

Related papers

• Query Drift Compensation: Enabling Compatibility In Continual Learning Of Retrieval Embedding Models (2025)1.20
• Align Then Train: Efficient Retrieval Adapter Learning (2026)0.00
• Federated Learning With Ad-hoc Adapter Insertions: The Case Of Soft-embeddings For Training Classifier-as-retriever (2025)0.00
• Matryoshka-adaptor: Unsupervised And Supervised Tuning For Smaller Embedding Dimensions (2024)2.26
• Gleanvec: Accelerating Vector Search With Minimalist Nonlinear Dimensionality Reduction (2024)0.00
• Parameter-efficient Sparse Retrievers And Rerankers Using Adapters (2023)4.52
• Progressively Optimized Bi-granular Document Representation For Scalable Embedding Based Retrieval (2022)11.06
• PEFA: Parameter-free Adapters For Large-scale Embedding-based Retrieval Models (2023)7.73