ARMED: Adaptive Retrieval-Augmented Medication Recommendation

Kyungkeon Chung; Se Un Park; Hyejin Lee; Junyoung Park

cs.LGOct 12, 2026

ARMED: Adaptive Retrieval-Augmented Medication Recommendation

Authors: Kyungkeon Chung, Se Un Park, Hyejin Lee, Junyoung Park

Organizations: UX Factory, Inc., Seongnam-si, South Korea

Abstract

Medication recommendation requires jointly considering diagnoses, comorbidities, longitudinal patient history, and treatment interactions. Although recent retrieval-augmented generation approaches using language models have shown promise for clinical recommendation, most existing systems rely on fixed retrieval pipelines that cannot adapt to diverse clinical retrieval scenarios and may introduce noisy or irrelevant evidence. We propose ARMED, an adaptive clinical RAG framework that dynamically expands retrieval according to retrieval sufficiency and current-admission relevance. ARMED first observes the initial retrieval outcome and selectively performs supplementary or fallback multi-query retrieval when additional evidence is needed. The framework also incorporates a relevance gate that activates only clinically relevant same-patient history and organizes retrieved evidence into structured query-level clinical evidence. Experiments on MIMIC-IV demonstrate that ARMED consistently outperforms previous static retrieval baselines and improves overall medication recommendation quality. By combining locally runnable small language models with adaptive retrieval, ARMED supports medication recommendation without relying on external language model APIs or task-specific fine-tuning.

Figures & tables

Fig. 3: Adaptive retrieval with single-query selection, multi-query expansion, rewriting, and rescue.

Method	F1	95% CI	Precision	Recall
History	0.2926	[0.255, 0.326]	0.2859	0.3352
Static RAG	0.4093	[0.397, 0.422]	0.3841	0.5084
Static RAG + History	0.4472	[0.432, 0.463]	0.4121	0.5669
TreatRAG	0.4009	[0.378, 0.425]	0.4002	0.4712
PACE-RAG	0.3557	[0.312, 0.397]	0.4157	0.3553
ARMED	0.5103	[0.491, 0.531]	0.4910	0.6263

TABLE I: Overall medication recommendation performance.

Comparison	$\Delta$ F1	95% CI	$p_{\mathrm{Holm}}$
vs. Static RAG + History	+0.0631	[0.0522, 0.0735]	$<0.001$
vs. TreatRAG	+0.1094	[0.0916, 0.1279]	$<0.001$
vs. PACE-RAG	+0.1546	[0.1271, 0.1855]	$<0.001$

TABLE II: Statistical significance of F1 improvements.

Method	Predicted	TP	FP	FN
Static RAG + History	15.834	6.490	9.344	6.136
ARMED	14.652	7.146	7.506	5.480

TABLE III: Mean medication prediction counts per admission.

Method	F1	Precision	Recall
w/o History	0.4387	0.4279	0.5293
Always-History	0.4783	0.4629	0.5757
ARMED	0.4792	0.4635	0.5777

TABLE IV: Effect of same-patient history selection.

Method	F1	GT Coverage	Evidence Precision
Single-query only	0.4178	0.7174	0.2406
Multi-query only	0.3943	0.6329	0.2690
Adaptive (ARMED)	0.4417	0.7825	0.2437

TABLE V: Retrieval policy ablation without same-patient history.

Metric	Adaptive	Multi-query only
F1	0.4417	0.3943
Mean latency (s)	14.686	16.865
P50 latency (s)	12.088	13.466
P95 latency (s)	28.630	33.934
Throughput (/h)	245.1	213.5

TABLE VI: Performance and runtime comparison without same-patient history.

Route	$N$	F1	Precision	Recall
Single-query sufficient	449	0.4787	0.4587	0.5881
Supplement	448	0.4167	0.4215	0.4962
Fallback	103	0.3888	0.3884	0.4798

TABLE VII: Retrieval strategy breakdown without same-patient history.

Backbone	Static RAG + History	ARMED	$\Delta$ F1
MedGemma 1.5 4B	0.4074	0.4891	+0.0817
Qwen3-8B	0.4472	0.5103	+0.0631
Phi-3.5 Mini 3.8B	0.3687	0.4300	+0.0613
Gemma 3 4B	0.4323	0.4825	+0.0502
EXAONE 3.5 2.4B	0.4143	0.4554	+0.0411
Ministral 3 8B	0.4900	0.4921	+0.0021

TABLE VIII: Medication recommendation performance across backbone models.

Case Type	Ambiguous Comorbidity Case	Weak Retrieval Case	History Noise Control Case
Clinical Description	Advanced pulmonary malignancy with respiratory failure, renal disease, and cardiac ischemic comorbidity	End-stage renal disease with metabolic derangements and complex renal/urologic history	Neurologic/vascular presentation with diabetes and cardiovascular risk factors, but weakly relevant prior history
Baseline Failure	Broad retrieval caused many FP/FN errors	Single-query evidence failed to recover sufficient treatments	Recent history context increased FP burden
ARMED Action	Added multi-query supplement and retained relevant history	Switched to multi-query fallback	Rejected prior visits via history gate
Evidence Change	Recovered focus-specific candidates	Retrieved diagnosis-focused evidence	Suppressed noisy longitudinal context
F1 Change	0.125 $\rightarrow$ 0.667	0.400 $\rightarrow$ 0.812	0.235 $\rightarrow$ 0.500

TABLE IX: Case studies of static retrieval failure and adaptive recovery.

Method	F1	95% CI	Precision	Recall
Static RAG + History	0.4631	[0.4590, 0.4671]	0.4144	0.6179
ARMED	0.4878	[0.4836, 0.4918]	0.4601	0.6189

TABLE X: Medication recommendation performance on MIMIC-III.

Explore similar work

PACE-RAG: Patient-Aware Contextual and Evidence-Constrained RAG for Clinical Drug Recommendation
Mar 18, 2026Chaeyoung Huh, Hyunmin Hwang, Jung Hwan Shin +3Retrieval-Augmented Generation Clinical Decision Support
1Korea Advanced Institute of Science and Technology, KAIST, Republic of Korea · Department of Neurology, Seoul National University Hospital, Republic of Korea · 3Asan Medical Center, University of Ulsan, College of Medicine, Republic of Korea +1
HypeMed: Enhancing Medication Recommendations with Hypergraph-Based Patient Relationships
Mar 19, 2026Xiangxu Zhang, Xiao Zhou, Hongteng Xu +1Recommender Systems Clinical Decision Support
Gaoling School of Artificial Intelligence, Renmin University of China, China · Beijing Key Laboratory of Research on Large Models and Intelligent Governance · Engineering Research Center of Next-Generation Intelligent Search and Recommendation, MOE +1
SafeRx-Agent: A Knowledge-Grounded Multi-Agent Framework for Safe and Explainable Medication Recommendation
May 27, 2026Xinyu Wang, Hanwei Wu, Zhenghan Tai +7Multi-Agent LLM Systems Healthcare
McGill University · McMaster University · University of Toronto +1

Mar 18, 2026J/K move · Enter open · S save

PACE-RAG: Patient-Aware Contextual and Evidence-Constrained RAG for Clinical Drug Recommendation

Chaeyoung Huh, Hyunmin Hwang, Jung Hwan Shin, Sungyang Jo+2

1Korea Advanced Institute of Science and Technology, KAIST, Republic of Korea · Department of Neurology, Seoul National University Hospital, Republic of Korea · 3Asan Medical Center, University of Ulsan, College of Medicine, Republic of Korea+1

Retrieval-Augmented Generation Clinical Decision Support Parkinson's Disease

Open paper AskPDF arXiv

Drug recommendation requires a deep understanding of individual patient context, especially for complex conditions like Parkinson's disease. While LLMs possess broad medical knowledge, they fail to capture the subtle nuances of actual prescribing patterns. Existing RAG methods also struggle with these complexities because guideline-based retrieval remains too generic and similar-patient retrieval often replicates majority patterns without accounting for the unique clinical nuances of individual patients. To bridge this gap, we propose PACE-RAG (Patient-Aware Contextual and Evidence-Constrained RAG). Rather than directly copying frequent medications from retrieved patients, PACE-RAG personalizes recommendations by first extracting patient-specific clinical features, retrieving cases around these features, and then refining the final prescription using the patient's current symptoms, active medication history, and focus-specific prescribing tendencies. By analyzing treatment patterns tailored to specific clinical features, PACE-RAG generates patient-specific medication recommendations along with an explainable clinical summary. PACE-RAG achieved the strongest performance among the evaluated inference-only LLM-based methods, reaching F1 scores of 80.84% and 47.22% on the Parkinson's disease and MIMIC-IV cohorts, respectively. Our code is available at: https://github.com/ChaeYoungHuh/PACE-RAG.

Abstract

Explore similar work

PACE-RAG: Patient-Aware Contextual and Evidence-Constrained RAG for Clinical Drug Recommendation

HypeMed: Enhancing Medication Recommendations with Hypergraph-Based Patient Relationships

SafeRx-Agent: A Knowledge-Grounded Multi-Agent Framework for Safe and Explainable Medication Recommendation