# Drift-Adapter: A Practical Approach to Near Zero-Downtime Embedding Model Upgrades in Vector Databases

arXiv: https://arxiv.org/abs/2509.23471 · HTML: https://mlsanity.com/paper/2509.23471/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 6 of its 16 references that are in the map.
- First posted on arXiv 27 Sep 2025.

## Builds on (most cited first, 6 of 6)

- [Billion-scale similarity search with GPUs](https://mlsanity.com/paper/1702.08734/) (2017)
- [LAION-400M: Open Dataset of CLIP-Filtered 400 Million Image-Text Pairs](https://mlsanity.com/paper/2111.02114/) (2021)
- [MTEB: Massive Text Embedding Benchmark](https://mlsanity.com/paper/2210.07316/) (2022)
- [Memory-Efficient Incremental Learning Through Feature Adaptation](https://mlsanity.com/paper/2004.00713/) (2020)
- [SPFresh: Incremental In-Place Update for Billion-Scale Vector Search](https://mlsanity.com/paper/2410.14452/) (2024)
- [Vec2Vec: A Compact Neural Network Approach for Transforming Text Embeddings with High Fidelity](https://mlsanity.com/paper/2306.12689/) (2023)

## Built on by (most cited first, 3 of 3)

- [Align Then Adapt: Label-Efficient Adapter Learning for Asymmetric Dense Retrieval](https://mlsanity.com/paper/2604.03403/) (2026)
- [Exploiting Residual Reachability for Cross-Model Migration of Graph-Based Indexes in Approximate Nearest Neighbor Search](https://mlsanity.com/paper/2609.23036/) (2026)
- [ROSETTA: Rotation Subspace Estimation for Training-free Transmodal Alignment](https://mlsanity.com/paper/2610.12871/) (2026)

## Abstract (arXiv)

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 like full re-indexing or dual-index serving, Drift-Adapter reduces recompute costs by over 100 times and facilitates upgrades with near-zero operational interruption. We analyze robustness to varied model drift, training data size, scalability to billion-item systems, and the impact of design choices like diagonal scaling, demonstrating Drift-Adapter's viability as a pragmatic solution for agile model deployment.