Automated Medical Information Retrieval via Agentic Filter-and-Refine Pipeline

Efficiency Analysis of Fine-Tuning Compact Language Models for Biomedical Information Retrieval Systems

Enhancing medical Q&A systems with multimodal knowledge graphs and dual-layer attention mechanisms

This study develops a text-based intent recognition model with a dual-layer attention architecture, in which a global contextual attention module is introduced to capture long-range semantic dependencies and improve multi-label classification performance.

A Novel Approach Based on Integrating Small Language Models and Retrieval-Augmented Generation for Medical Question Answering

—This paper proposes a novel approach which combines small language models with retrieval-augmented generation in medical question answering to provide accurate and comprehensible information. Our method extracts relevant evidence from external knowledge and converts them into vector embeddings which are used for high-fidelity semantic retrieval. The small language model subsequently synthesizes the retrieved evidence into fluent, context-aware responses. Extensive experiments conducted on the PubMed benchmark dataset, the results show competitive performance to larger language models while being far more suitable for deployment in resource-constrained environments. Moreover, the proposed method supports efficient domain knowledge updates without the need for extensive re-training. Our implementation is available in the following GitHub repository: https://github.com/LeoBaoNguyen12/RagSLM-MQA.

An Enhanced Topic Extraction Model for Medical PubMed Documents using State-of-the-Art Algorithms

The rapid growth of medical literature databases represents both a challenge and an opportunity for pharmacovigilance. Medical abstracts are full of specialized terms and complex sentences that make extracting meaningful insights on the adverse effects of drugs very challenging. This paper addresses the critical problem of extracting relevant topics related to drug adverse effects from PubMed medical abstracts using advanced topic modelling methods. It enhances the four-topic modelling with two optimization algorithms to improve topic extraction and assesses their performance, such as Latent Dirichlet Allocation (LDA), Latent Semantic Analysis (LSA), Long Short-Term Memory (LSTM), and Recurrent Neural Network (RNN), combined with grid search and Bayesian optimization algorithms. The experimental results show that LDA optimized with Bayesian optimization gives the highest coherence score, 0.605, which is better than other models. Coherent results, as shown in a complex comparison table, reveal the performance of each model and optimization method.

Neighborhood-Aware Dual Biomedical Entity Linking

Biomedical entity linking grounds mentions in clinical and scientific text to entities in a curated knowledge base (KB) with ontological structure, which supports downstream applications such as literature-scale information extraction and patient-record normalization. The task has several challenges at once: the KB contains large numbers of entities, mentions are often ambiguous, and gold labels follow annotation conventions specific to each corpus. To address these challenges, we propose PILOT, a three-stage framework made up of neighborhood-aware retrieval, dual reranking, and score fusion. The retriever injects ontological structure from both the query and KB side, by reformulating mentions and pooling entity embeddings. The retrieved pool is then scored from two complementary views, one over surface forms and one over context, and fused together. PILOT achieves the state of the art on average across five widely-used benchmarks and remains efficient at inference.

A Hybrid Retrieval System for Adverse Event Concept Normalization Integrating Contextual Scoring, Lexical Augmentation, and Semantic Fine-Tuning