# Embedding Surgery: Localized Updates for Adaptive Ranking Correction in Dense Retrieval

arXiv: https://arxiv.org/abs/2609.05110 · HTML: https://mlsanity.com/paper/2609.05110/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Generation and context integration, one of 15 directions in Retrieval-augmented generation.
- Builds on 23 of its 46 references that are in the map.
- First posted on arXiv 4 Sep 2026.

## Placed in

- Retrieval-augmented generation / [Generation and context integration](https://mlsanity.com/rag/generation-and-context-integration/)

## Builds on (most cited first, 23 of 23)

- [Billion-scale similarity search with GPUs](https://mlsanity.com/paper/1702.08734/) (2017)
- [Efficient and robust approximate nearest neighbor search using Hierarchical Navigable Small World graphs](https://mlsanity.com/paper/1603.09320/) (2016)
- [CVXPY: A Python-Embedded Modeling Language for Convex Optimization](https://mlsanity.com/paper/1603.00943/) (2016)
- [Overview of the TREC 2019 deep learning track](https://mlsanity.com/paper/2003.07820/) (2020)
- [Efficiently Teaching an Effective Dense Retriever with Balanced Topic Aware Sampling](https://mlsanity.com/paper/2104.06967/) (2021)
- [Overview of the TREC 2020 deep learning track](https://mlsanity.com/paper/2102.07662/) (2021)
- [Unbiased Learning-to-Rank with Biased Feedback](https://mlsanity.com/paper/1608.04468/) (2016)
- [Fast Approximate Nearest Neighbor Search With The Navigating Spreading-out Graph](https://mlsanity.com/paper/1707.00143/) (2017)
- [Perspectives on Large Language Models for Relevance Judgment](https://mlsanity.com/paper/2304.09161/) (2023)
- [Arctic-Embed 2.0: Multilingual Retrieval Without Compromise](https://mlsanity.com/paper/2412.04506/) (2024)
- [LLM-based relevance assessment still can't replace human relevance assessment](https://mlsanity.com/paper/2412.17156/) (2024)
- [Differentiable Unbiased Online Learning to Rank](https://mlsanity.com/paper/1809.08415/) (2018)
- [How Deep is your Learning: the DL-HARD Annotated Deep Learning Dataset](https://mlsanity.com/paper/2105.07975/) (2021)
- [Injecting the BM25 Score as Text Improves BERT-Based Re-rankers](https://mlsanity.com/paper/2301.09728/) (2023)
- [Unifying Online and Counterfactual Learning to Rank](https://mlsanity.com/paper/2012.04426/) (2020)
- [A Human-AI Comparative Analysis of Prompt Sensitivity in LLM-Based Relevance Judgment](https://mlsanity.com/paper/2504.12408/) (2025)
- [To Model or to Intervene: A Comparison of Counterfactual and Online Learning to Rank from User Interactions](https://mlsanity.com/paper/1907.06412/) (2019)
- [Studying Catastrophic Forgetting in Neural Ranking Models](https://mlsanity.com/paper/2101.06984/) (2021)
- [Caching Historical Embeddings in Conversational Search](https://mlsanity.com/paper/2211.14155/) (2022)
- [Efficient Neural Ranking using Forward Indexes and Lightweight Encoders](https://mlsanity.com/paper/2311.01263/) (2023)
- [Reinforcement Online Learning to Rank with Unbiased Reward Shaping](https://mlsanity.com/paper/2201.01534/) (2022)
- [Implicit Feedback for Dense Passage Retrieval: A Counterfactual Approach](https://mlsanity.com/paper/2204.00718/) (2022)
- [When Should Dense Retrievers Be Updated in Evolving Corpora? Detecting Out-of-Distribution Corpora Using GradNormIR](https://mlsanity.com/paper/2506.01877/) (2025)

## Abstract (arXiv)

Dense retrieval systems are core components of modern search engines, recommendation platforms, and retrieval-augmented generation pipelines. They encode documents and queries into dense embeddings, enabling efficient semantic search via vector similarity. However, because document embeddings are computed offline and stored in static indexes, these systems struggle to adapt to user feedback or evolving search intent. To address this limitation, we introduce embedding surgery, a lightweight approach for adaptive ranking correction in dense retrieval. The method applies localized, minimal updates to selected document embeddings at query time, guided by editorial feedback, user interactions, or pseudo-labels from large language models. We formulate embedding surgery as a convex optimization problem that enforces ranking constraints while minimizing modifications to the affected document representations. We integrate embedding surgery into standard dense retrieval pipelines and evaluate it on TREC Deep Learning, TREC Robust, TREC CAsT, and MS MARCO benchmarks. Results show consistent improvements (e.g., up to +60.64% relative improvement in nDCG@10 on DL-Hard under editorial feedback), even under noisy or shifting feedback, with low computational cost and without disrupting the global structure of the embedding space. Extensive experiments show that ranking corrections propagate to semantically related queries and that embedding updates can be applied safely and efficiently to scalable Approximate Nearest Neighbor indexes via simple in-place overwriting, without requiring costly index reconstruction. Finally, embedding surgery complements query adaptation methods such as CoRocchio, yielding additional gains while being more robust to noisy feedback.