The integration of databases and machine learning promises to enhance various aspects of data management, analysis, and application.However, in-database machine learning (In-DB ML) is not easily portable to different database management systems and current approaches are typically limited to training and inference, while modern machine learning pipelines often involve aspects such as continuous learning, unlearning, and explainability.This paper presents BornSQL, a In-DB ML algorithm based on the Born Classifier [7], and exclusively implemented through standard SQL queries.BornSQL can handle categorical data, and it is particularly appropriate for classification of textual data.Further contributions of BornSQL are i) incremental learning to efficiently enforce model updates when new data become available in the db, ii) unlearning when selected data needs to be excluded due to privacy issues, and iii) global/local explainability to associate the importance of a feature/attribute in determining the classification result.We illustrate the usage and scalability of the algorithm using a benchmark database consisting of 2,359,828 scientific publications divided into three classes and composed of 3,942,559 features.The training time is linear in the number of publications and the average inference time for a publication is 1 millisecond on our experimental environment.We discuss potential applications such as cost-effective model serving, exploratory data analysis, and data privacy.

Despite recent progress, Biomedical Entity Linking (BEL) with large language models (LLMs) remains computationally inefficient and challenging to deploy in practical settings. In this work, we demonstrate that instruction-tuning of open-source generative models can offer an effective solution when applied at the re-ranking stage of the BEL pipeline. We propose a set-wise instruction-tuning formulation that enables fast and accurate candidate selection. Our method demonstrates strong performance on multiple BEL benchmarks, yielding significant improvements in linking accuracy (3%–24%) while reducing inference time compared to the state-of-the-art. We integrate our generative re-ranker into BeLink, a modular, end-to-end system designed for practical real-world BEL applications.

MOTIVATION: Biomedical Entity Linking (BEL) maps mentions in biomedical text to standardized identifiers, enabling structured data integration and downstream knowledge discovery. However, current BEL systems remain fundamentally constrained by the recall of the initial candidate pool, where suboptimal retrieval limits the overall effectiveness of the normalization pipeline. RESULTS: We present the first systematic evaluation of Generative Relevance Feedback (GRF) for enhancing candidate retrieval in state-of-the-art BEL systems. GRF leverages large language models (LLMs) to enrich the expressiveness of the mention in a zero-shot fashion. We assess GRF's impact under two scenarios-direct linking prediction and candidate generation in cascading normalization pipelines-and analyze its sensitivity to different LLMs, feedback types, and integration strategies. Experiments across eight corpora and four biomedical knowledge bases demonstrate that integrating GRF significantly improves both accuracy and recall, thereby increasing the upper bound on normalization performance. Our findings highlight GRF as an efficient, model-agnostic solution and underscore its potential as a key component for advancing BEL. AVAILABILITY AND IMPLEMENTATION: The code to reproduce our experiments can be found at: https://doi.org/10.5281/zenodo.17853541.