A Human-Centric Pipeline for Aligning Large Language Models with Chinese Medical Ethics

2026-01-12

AI Summary

This study addresses the challenge of aligning large language models with China-specific ethical norms in complex medical scenarios. To this end, the authors construct MedES, a dynamic scenario benchmark grounded in 260 authoritative medical ethics and legal provisions, and propose a “Guardian-in-the-Loop” framework coupled with a modular, replaceable normative corpus to enable a scalable alignment paradigm for Chinese medical ethics. Leveraging expert-annotated data, they train a high-precision automatic evaluator and apply supervised fine-tuning combined with domain-specific preference optimization to ethically align a 7B-parameter model. The aligned model significantly outperforms larger-scale baselines on Chinese medical ethics tasks, demonstrating marked improvements in both quality and composite evaluation metrics.

Technology Category

• Natural Language Processing: Ethics — Bias, Fairness, Transparency & Privacy
• Philosophy and Ethics of AI: Bias, Fairness & Equity
• Machine Learning: Ethics, Bias, and Fairness

Application Category

• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization
• Economics, Online Markets and Human Computation: Fairness and ethical considerations in crowd work and in human-in-the-loop AI systems
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Recent advances in large language models have enabled their application to a range of healthcare tasks. However, aligning LLMs with the nuanced demands of medical ethics, especially under complex real world scenarios, remains underexplored. In this work, we present MedES, a dynamic, scenario-centric benchmark specifically constructed from 260 authoritative Chinese medical, ethical, and legal sources to reflect the challenges in clinical decision-making. To facilitate model alignment, we introduce a guardian-in-the-loop framework that leverages a dedicated automated evaluator (trained on expert-labeled data and achieving over 97% accuracy within our domain) to generate targeted prompts and provide structured ethical feedback. Using this pipeline, we align a 7B-parameter LLM through supervised fine-tuning and domain-specific preference optimization. Experimental results, conducted entirely within the Chinese medical ethics context, demonstrate that our aligned model outperforms notably larger baselines on core ethical tasks, with observed improvements in both quality and composite evaluation metrics. Our work offers a practical and adaptable framework for aligning LLMs with medical ethics in the Chinese healthcare domain, and suggests that similar alignment pipelines may be instantiated in other legal and cultural environments through modular replacement of the underlying normative corpus.

Problem

• medical ethics
• large language models
• Chinese healthcare
• clinical decision-making
• ethical alignment

Innovation

• medical ethics alignment
• guardian-in-the-loop
• scenario-centric benchmark
• domain-specific preference optimization
• Chinese healthcare LLM