{"slug": "refine-llm-refinement-over-budgeted-text-attributed-graphs-for-personalized", "title": "REFINE: LLM Refinement over Budgeted Text-Attributed Graphs for Personalized Medical Concept Representation", "summary": "Researchers propose REFINE, a KG-aware budgeted LLM graph refinement framework that learns patient-personalized medical concept representations from text-attributed knowledge graphs, using a sequential reinforcement learning policy to select a personalized KG expansion budget for each observed code and a frozen LLM with graph-aware soft prompts for semantic refinement. Experiments on MIMIC-III and MIMIC-IV show REFINE consistently improves diverse EHR backbones and outperforms strong baselines across component ablation, KG selection, and data insufficiency.", "body_md": "arXiv:2609.04415v1 Announce Type: new \nAbstract: Learning rich medical concept representations is essential for EHR prediction. Text-attributed knowledge graphs (TKGs) provide a natural foundation by organizing heterogeneous medical relations together with textual semantics. However, most existing encoders process concepts uniformly across patients, despite the fact that a code's meaning and predictive value depend on patient-specific clinical context and trajectory. Learning patient-personalized concept representations from TKGs introduces two key challenges: (1) deciding how much KG context to incorporate for each observed code, and (2) aligning semantic information with the patient-specific relational structure. We propose REFINE, a KG-aware budgeted LLM graph refinement framework for patient-personalized medical concept encoding. Starting from a global TKG, REFINE constructs patient-specific temporal graphs. A sequential reinforcement learning policy selects a personalized KG expansion budget for each observed code. The resulting patient graph is processed by a heterogeneous GNN to capture relation-aware structural dependencies, while a frozen LLM uses graph-aware soft prompts to semantically refine concept representations. Experiments on MIMIC-III and MIMIC-IV show that REFINE consistently improves diverse EHR backbones, outperforms strong baselines, and demonstrates robust gains across component ablation, KG selection, and data insufficiency.", "url": "https://wpnews.pro/news/refine-llm-refinement-over-budgeted-text-attributed-graphs-for-personalized", "canonical_source": "https://arxiv.org/abs/2609.04415", "published_at": "2026-09-07 04:00:00+00:00", "updated_at": "2026-09-07 04:27:10.724580+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "machine-learning"], "entities": ["REFINE", "MIMIC-III", "MIMIC-IV"], "alternates": {"html": "https://wpnews.pro/news/refine-llm-refinement-over-budgeted-text-attributed-graphs-for-personalized", "markdown": "https://wpnews.pro/news/refine-llm-refinement-over-budgeted-text-attributed-graphs-for-personalized.md", "text": "https://wpnews.pro/news/refine-llm-refinement-over-budgeted-text-attributed-graphs-for-personalized.txt", "jsonld": "https://wpnews.pro/news/refine-llm-refinement-over-budgeted-text-attributed-graphs-for-personalized.jsonld"}}