Fine-grained List-wise Alignment for Generative Medication Recommendation

• URL: http://arxiv.org/abs/2505.20218v1
• Date: Mon, 26 May 2025 16:59:23 GMT
• Title: Fine-grained List-wise Alignment for Generative Medication Recommendation
• Authors: Chenxiao Fan, Chongming Gao, Wentao Shi, Yaxin Gong, Zihao Zhao, Fuli Feng,
• Abstract summary: We propose FLAME, a fine-grained list-wise alignment framework for large language models (LLMs)<n> FLAME formulates recommendation as a sequential decision process, where each step adds or removes a single drug.<n> Experiments on benchmark datasets demonstrate that FLAME achieves state-of-the-art performance.
• Score: 30.397147691681607
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Accurate and safe medication recommendations are critical for effective clinical decision-making, especially in multimorbidity cases. However, existing systems rely on point-wise prediction paradigms that overlook synergistic drug effects and potential adverse drug-drug interactions (DDIs). We propose FLAME, a fine-grained list-wise alignment framework for large language models (LLMs), enabling drug-by-drug generation of drug lists. FLAME formulates recommendation as a sequential decision process, where each step adds or removes a single drug. To provide fine-grained learning signals, we devise step-wise Group Relative Policy Optimization (GRPO) with potential-based reward shaping, which explicitly models DDIs and optimizes the contribution of each drug to the overall prescription. Furthermore, FLAME enhances patient modeling by integrating structured clinical knowledge and collaborative information into the representation space of LLMs. Experiments on benchmark datasets demonstrate that FLAME achieves state-of-the-art performance, delivering superior accuracy, controllable safety-accuracy trade-offs, and strong generalization across diverse clinical scenarios. Our code is available at https://github.com/cxfann/Flame.

Related papers

• Combating the Bucket Effect:Multi-Knowledge Alignment for Medication Recommendation [7.962835256943926]
  We introduce a cross-modal medication encoder capable of seamlessly aligning data from different modalities.<n>We then combine the multi-knowledge medication representations with patient records for recommendations.<n>Experiments show that MKMed mitigates the "bucket effect" in data, and significantly outperforms state-of-the-art baselines in recommendation accuracy and safety.
  arXiv  Detail & Related papers  (2025-04-25T05:47:15Z)
• Fine-grained Alignment of Large Language Models for General Medication Recommendation without Overprescription [45.41664696802343]
  Large language models (LLMs) hold significant promise in achieving general medication recommendation systems.<n>We introduce Language-Assisted Medication Recommendation, which tailors LLMs for medication recommendation in a medication-aware manner.<n>Fine-tuning LLMs with this framework can outperform existing methods by more than 10% in internal validation and generalize across temporal and external validations.
  arXiv  Detail & Related papers  (2025-03-05T17:28:16Z)
• DrugImproverGPT: A Large Language Model for Drug Optimization with Fine-Tuning via Structured Policy Optimization [53.27954325490941]
  Finetuning a Large Language Model (LLM) is crucial for generating results towards specific objectives.<n>This research introduces a novel reinforcement learning algorithm to finetune a drug optimization LLM-based generative model.
  arXiv  Detail & Related papers  (2025-02-11T04:00:21Z)
• ADEP: A Novel Approach Based on Discriminator-Enhanced Encoder-Decoder Architecture for Accurate Prediction of Adverse Effects in Polypharmacy [0.0]
  We introduce ADEP, a novel approach integrating a discriminator and an encoder-decoder model to address data sparsity and enhance feature extraction. ADEP employs a three-part model, including multiple classification methods, to predict adverse effects in polypharmacy.
  arXiv  Detail & Related papers  (2024-05-31T18:20:17Z)
• CIDGMed: Causal Inference-Driven Medication Recommendation with Enhanced Dual-Granularity Learning [10.60553153370577]
  Medication recommendation aims to integrate patients' long-term health records to provide accurate and safe medication combinations. Existing methods often fail to deeply explore the true causal relationships between diseases/procedures and medications. We propose the Causal Inference-driven Dual-Granularity Medication Recommendation method (CIDGMed)
  arXiv  Detail & Related papers  (2024-03-01T08:50:27Z)
• Large Language Model Distilling Medication Recommendation Model [58.94186280631342]
  We harness the powerful semantic comprehension and input-agnostic characteristics of Large Language Models (LLMs)<n>Our research aims to transform existing medication recommendation methodologies using LLMs.<n>To mitigate this, we have developed a feature-level knowledge distillation technique, which transfers the LLM's proficiency to a more compact model.
  arXiv  Detail & Related papers  (2024-02-05T08:25:22Z)
• Zero-shot Learning of Drug Response Prediction for Preclinical Drug Screening [38.94493676651818]
  We propose a zero-shot learning solution for the. task in preclinical drug screening. Specifically, we propose a Multi-branch Multi-Source Domain Adaptation Test Enhancement Plug-in, called MSDA.
  arXiv  Detail & Related papers  (2023-10-05T05:55:41Z)
• SSM-DTA: Breaking the Barriers of Data Scarcity in Drug-Target Affinity Prediction [127.43571146741984]
  Drug-Target Affinity (DTA) is of vital importance in early-stage drug discovery. wet experiments remain the most reliable method, but they are time-consuming and resource-intensive. Existing methods have primarily focused on developing techniques based on the available DTA data, without adequately addressing the data scarcity issue. We present the SSM-DTA framework, which incorporates three simple yet highly effective strategies.
  arXiv  Detail & Related papers  (2022-06-20T14:53:25Z)
• SafeDrug: Dual Molecular Graph Encoders for Safe Drug Recommendations [59.590084937600764]
  We propose a DDI-controllable drug recommendation model named SafeDrug to leverage drugs' molecule structures and model DDIs explicitly. On a benchmark dataset, our SafeDrug is relatively shown to reduce DDI by 19.43% and improves 2.88% on Jaccard similarity between recommended and actually prescribed drug combinations over previous approaches.
  arXiv  Detail & Related papers  (2021-05-05T00:20:48Z)
• SDF-Bayes: Cautious Optimism in Safe Dose-Finding Clinical Trials with Drug Combinations and Heterogeneous Patient Groups [84.63561578944183]
  This paper proposes a novel Bayesian design, SDF-Bayes, for finding the maximum tolerated dose (MTD) of a drug in a clinical trial. Rather than the conventional principle of escalating or de-escalating the current dose of one drug, SDF-Bayes proceeds by cautious optimism. Extensive experiments based on both synthetic and real-world datasets demonstrate the advantages of SDF-Bayes over state of the art DC trial designs.
  arXiv  Detail & Related papers  (2021-01-26T18:59:26Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.