Enhancing the Medical Context-Awareness Ability of LLMs via Multifaceted Self-Refinement Learning

2025/11/13 by Zhou, Yuxuan, Wang, Yubin, Wang, Bin +4
Computer Science · Medicine · #Artificial Intelligence (cs.AI) #Artificial Intelligence in Healthcare and Education #Computation and Language (cs.CL) #FOS: Computer and information sciences #Machine Learning in Healthcare #Topic Modeling

paper · doi:10.48550/arxiv.2511.10067

openalex publication_date 2025/11/13 · openalex created_date 2025/11/15 · openalex updated_date 2026/07/28

Abstract

Large language models (LLMs) have shown great promise in the medical domain, achieving strong performance on several benchmarks. However, they continue to underperform in real-world medical scenarios, which often demand stronger context-awareness, i.e., the ability to recognize missing or critical details (e.g., user identity, medical history, risk factors) and provide safe, helpful, and contextually appropriate responses. To address this issue, we propose Multifaceted Self-Refinement (MuSeR), a data-driven approach that enhances LLMs' context-awareness along three key facets (decision-making, communication, and safety) through self-evaluation and refinement. Specifically, we first design a attribute-conditioned query generator that simulates diverse real-world user contexts by varying attributes such as role, geographic region, intent, and degree of information ambiguity. An LLM then responds to these queries, self-evaluates its answers along three key facets, and refines its responses to better align with the requirements of each facet. Finally, the queries and refined responses are used for supervised fine-tuning to reinforce the model's context-awareness ability. Evaluation results on the latest HealthBench dataset demonstrate that our method significantly improves LLM performance across multiple aspects, with particularly notable gains in the context-awareness axis. Furthermore, by incorporating knowledge distillation with the proposed method, the performance of a smaller backbone LLM (e.g., Qwen3-32B) surpasses its teacher model, achieving a new SOTA across all open-source LLMs on HealthBench (63.8%) and its hard subset (43.1%). Code and dataset will be released at https://muser-llm.github.io.

Citations

• Baichuan-M2: Scaling Medical Capability with Large Verifier System
• Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
• Pangu Embedded: An Efficient Dual-system LLM Reasoner with Metacognition
• A Modular Approach for Clinical SLMs Driven by Synthetic Data with Pre-Instruction Tuning, Model Merging, and Clinical-Tasks Alignment
• Qwen3 Technical Report
• HealthBench: Evaluating Large Language Models Towards Improved Human Health
• DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning
• Phi-4 Technical Report
• Give me Some Hard Questions: Synthetic Data Generation for Clinical QA
• Med42-v2: A Suite of Clinical LLMs
• Synthetic Patient-Physician Dialogue Generation from Clinical Notes Using LLM
• The Llama 3 Herd of Models
• Med42 -- Evaluating Fine-Tuning Strategies for Medical LLMs: Full-Parameter vs. Parameter-Efficient Approaches
• Towards Building Multilingual Language Model for Medicine
• MedBench: A Large-Scale Chinese Benchmark for Evaluating Medical Large Language Models
• Can Generalist Foundation Models Outcompete Special-Purpose Tuning? Case Study in Medicine
• MEDITRON-70B: Scaling Medical Pretraining for Large Language Models
• HuatuoGPT-II, One-stage Training for Medical Adaption of LLMs
• CMB: A Comprehensive Medical Benchmark in Chinese
• HuatuoGPT, towards Taming Language Model to Be a Doctor
• Clinical Camel: An Open Expert-Level Medical Language Model with Dialogue-Based Knowledge Encoding
• Towards Expert-Level Medical Question Answering with Large Language Models
• Capabilities of GPT-4 on Medical Challenge Problems
• GPT-4 Technical Report
• Large Language Models Encode Clinical Knowledge
• LiveQA: A Question Answering Dataset over Sports Live
• What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Exams
• DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter
• TinyBERT: Distilling BERT for Natural Language Understanding
• PubMedQA: A Dataset for Biomedical Research Question Answering
• HEAD-QA: A Healthcare Dataset for Complex Reasoning
• Distilling the Knowledge in a Neural Network

Cited by

• MedReflect: Teaching Medical LLMs to Self-Improve via Reflective Correction

Related

• MedReflect: Teaching Medical LLMs to Self-Improve via Reflective Correction
• Fine-Tuning Medical Language Models for Enhanced Long-Contextual Understanding and Domain Expertise
• MedInsight: A Multi-Source Context Augmentation Framework for Generating Patient-Centric Medical Responses using Large Language Models
• MedOdyssey: A Medical Domain Benchmark for Long Context Evaluation Up to 200K Tokens
• From Beginner to Expert: Modeling Medical Knowledge into General LLMs