Results demonstrate that InfoFlowEX equips LLMs with robust adaptability, achieving consistent gains over baselines with minimal task-specific customization, highlighting InfoFlowEX for real-world biomedical applications.
Abstract
Summary Biomedical information extraction (BIE) is fundamental for transforming unstructured biomedical text into structured, computable knowledge, yet the effectiveness of large language models (LLMs) remains limited by dataset heterogeneity and lack of unified benchmarks. We present InfoFlowEX, a unified framework for generalizable biomedical knowledge extraction with LLMs. InfoFlowEX incorporates an automated data integration pipeline using ontology-guided alignment to construct BIE-Corpus, a large-scale multi-domain benchmark unifying 40 public datasets for named entity recognition and relation extraction. We further introduce a task-conditioned schema instruction tuning strategy encoding 28 biomedical entity and relation types into a schema codebase, enabling LLMs to align heterogeneous annotations and generalize across settings. Finally, we evaluated InfoFlowEX in diverse applications, including evidence retrieval for question-answering, clinical diagnosis from electronic health records, and knowledge graph expansion. Results demonstrate that InfoFlowEX equips LLMs with robust adaptability, achieving consistent gains over baselines with minimal task-specific customization, highlighting InfoFlowEX for real-world biomedical applications.
A layered reliability framework is defined in which graph-based inference addresses knowledge incompleteness, retrieval-augmented prompt control mitigates instability, and ontology grounding reduces semantic ambiguity, providing a foundation for more reliable biomedical AI systems.
Biomedical named entity recognition (NER) and relation extraction (RE) remain challenging because biomedical texts contain ambiguous abbreviations, complex entity boundaries, domain-specific terminology, and implicit relations. This study proposes a prompt-enhanced and QLoRA-adapted large language model framework for biomedical information extraction. For NER, abbreviation-aware prompting supports candidate detection, contextual interpretation, boundary-aware generation, and schema-constrained outputs. For RE, entity markers identify a predefined target pair, while filtered UMLS and MeSH concepts provide concise evidence. DeepSeek-R1-Distill-Qwen-7B is adapted using LoRA over a 4-bit quantized frozen backbone. Experiments cover three NER and three RE datasets. Across three training seeds, the dataset-level macro-average F1 values are 0.909 ± 0.001 for NER and 0.787 ± 0.001 for RE. Seed-balanced paired bootstrap resampling with 10,000 aligned instance-level resamples confirms significant improvements over a matched deterministic simple-prompt baseline on all six datasets after Holm–Bonferroni correction (adjusted p < 0.001), with absolute F1 gains from +0.091 to +0.131. Repeated-run ablations show low variability and complementary contributions from task-structured prompting, knowledge filtering, deterministic validation, and parameter-efficient adaptation.
This work presents a scalable, reproducible framework for evaluating, optimizing, and interpreting LLMs for biomedical knowledge extraction, with a focus on gene–gene regulatory relation prediction, pathway component recognition, multimodal pathway figure understanding, and automated prompt optimization.
Assessment of five small, open-source LLMs in identifying semantic relationships between biomedical concepts confirms that direct fine-tuning effectively exceeds the reasoning bottlenecks of smaller LLMs, providing an accurate, automated methodology for the construction and evolution of specialised biomedical ontologies.
Tanay Aggarwal, Angelo Salatino, Francesco Osborne et al.· arXiv.org· 0 citations
Motivation Pharmacovigilance relies on accurate extraction of structured biomedical entities and their semantic relationships from scientific literature. However, most biomedical information extraction systems address named entity recognition (NER) and relation extraction as separate tasks trained on corpus-specific architectures, limiting scalability and cross-task knowledge sharing. Recent developments in instruction-tuned Large Language Models (LLMs) offer a promising alternative through unified generative extraction, but robust schema-grounded multitask adaptation for biomedical extraction is still understudied. Methods This study proposes a unified multitask instruction-tuned LLM framework that jointly performs biomedical NER and relation extraction across three benchmark corpora to identify chemical, disease, drug entities, as well as chemical-disease relations, drug-adverse event relations, and drug-drug interactions. Two general LLMs, Llama-3.2-3B-Instruct and Qwen3-8B, were fine-tuned using Low-Rank Adaptation (LoRA) under a shared generation interface that extracts both entity pairs and their underlying relation. Zero-shot and fine-tuned configurations were evaluated across all the tasks on their respective held-out test sets. Results Parameter-efficient fine-tuning substantially improved both entity and relation extraction performance across all tasks and model families. Fine-tuned Qwen3-8B achieved the strongest overall performance with 89.42% micro-averaged entity F1 and 62.32% micro-averaged relation F1. Fine-tuned Llama-3.2-3B achieved 87.63% entity F1 and 58.42% relation F1 despite its substantially smaller parameter count, outperforming the zero-shot 8B model on both tasks. Fine-tuning also reduced structured JSON parse failures from 23.5% to 0.11%, demonstrating stable schema internalization during supervised adaptation. Conclusion Schema-grounded multitask instruction tuning with LoRA provides a robust and computationally feasible framework for unified biomedical information extraction across heterogeneous benchmark corpora. The findings further demonstrate that schema-grounded adaptation is substantially more important than model scale alone for reliable extraction of structured biomedical relations. The gap between NER and relation extraction performance motivates future research on explicit negative-relation supervision and ontology-guided relation extraction.
Biomedical ontology normalization maps free-text expressions to standardized concepts, enabling consistent integration and analysis of biomedical data. This task remains challenging because lexical variation and subtle distinctions among hierarchically related concepts can obscure concept boundaries. We present OntologyAligner, a three-stage framework that combines ontology-aligned retrieval, large language model candidate reranking, and selective hierarchy-guided refinement. We also construct PhenoNormBench, a unified benchmark comprising 13,390 samples from seven Human Phenotype Ontology datasets. OntologyAligner achieved state-of-the-art performance on HPO normalization, with 88.78% Macro Top-1 Accuracy and 86.75% Micro Top-1 Accuracy, exceeding the strongest baseline by 4.85 and 5.07 percentage points, respectively. Ablation analyses showed complementary contributions from all three stages, and sensitivity analyses demonstrated stability across candidate-set sizes and model backbones. Applications to MONDO, MEDIC, and NCBITaxon further established portability to other ontologies. OntologyAligner offers a generalizable framework for accurate mapping of biomedical text to structured ontology concepts. PhenoNormBench and the code are publicly available at https://github.com/zhelishisongjie/OntologyAligner.
Jie Song, Zhichuan Xu, Ziyu Lu et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
