{"slug": "tabular-population-priors-as-structured-retrieval-for-ehr-medication-set", "title": "Tabular Population Priors as Structured Retrieval for EHR Medication Set Prediction", "summary": "Replacing PACERAG's patient-level retriever with a tabular population prior raised F1 from 0.303 to 0.338 (+11.5% relative) and Jaccard from 0.218 to 0.237 on 1,004 held-out MIMIC-IV admissions, according to an arXiv paper (2611.17341v1). The prior ranks medication class candidates from a sparse condition-medication co-occurrence table estimated from training admissions, with no additional LLM calls and no inference-time patient index. A non-LLM top-k baseline reached F1=0.248, confirming the prior is predictive before LLM integration.", "body_md": "arXiv:2611.17341v1 Announce Type: new \nAbstract: Medication prediction from electronic health records is a fixed vocabulary structured prediction problem, yet recent clinical RAG systems often treat it like open ended case based reasoning: retrieve similar patient admissions and ask an LLM to infer a medication set. This retrieval object is indirect the model receives retrieved patient records while the target is a discrete set of medication class labels. We propose a more task aligned retrieval object: a tabular population prior ranked medication class candidates from a sparse condition medication co occurrence table estimated from training admissions. Using PACERAG as a fixed downstream pipeline and replacing only its patient retriever, this drop in swap improves F1 from 0.303 to 0.338 (+11.5% relative) and Jaccard from 0.218 to 0.237 on 1,004 heldout MIMIC IV admissions, with no additional LLM calls and no inference time patient index. A non LLM top k baseline reaches F1=0.248, confirming the prior is predictive before LLM integration. Aggregate population statistics can provide a more task aligned, efficient, and auditable retrieval source than patient level nearest neighbors for fixed vocabulary EHR set prediction.", "url": "https://wpnews.pro/news/tabular-population-priors-as-structured-retrieval-for-ehr-medication-set", "canonical_source": "https://arxiv.org/abs/2611.17341", "published_at": "2026-11-16 05:00:00+00:00", "updated_at": "2026-11-16 05:19:36.362778+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "ai-research", "large-language-models", "ai-safety"], "entities": ["PACERAG", "MIMIC-IV", "arXiv"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/tabular-population-priors-as-structured-retrieval-for-ehr-medication-set", "markdown": "https://wpnews.pro/news/tabular-population-priors-as-structured-retrieval-for-ehr-medication-set.md", "text": "https://wpnews.pro/news/tabular-population-priors-as-structured-retrieval-for-ehr-medication-set.txt", "jsonld": "https://wpnews.pro/news/tabular-population-priors-as-structured-retrieval-for-ehr-medication-set.jsonld"}}