构建三层框架，让AI决策更透明可信。
A Multi-Layered Research Framework for Human-Centered AI: Defining the Path to Explainability and Trust
- 分三层设计：模型内置可解释性、解释适配用户认知、实时反馈优化
- 在医疗金融等领域提升决策质量与合规性
- 适合关注AI可信度与人机协同的研究者和开发者
人工智能在医疗、金融、自主系统等高风险领域应用常受限于透明度、可解释性和信任问题。尽管以人为本的AI（HCAI）强调与人类价值观对齐，可解释AI（XAI）通过增强决策透明性提升理解度，但缺乏统一方法限制了其在关键决策场景中的效能。本文提出一种新型三层框架，融合HCAI与XAI，建立结构化可解释性范式：第一层为具备内建可解释机制的基础AI模型；第二层为人本解释层，根据用户认知负荷与专业水平定制解释内容；第三层为动态反馈环，通过实时用户交互持续优化解释。该框架在医疗、金融及软件开发领域验证，证明其能有效提升决策质量、促进合规并增强公众信任。研究推动了以人为本的可解释AI（HCXAI）发展，助力构建透明、自适应且伦理对齐的AI系统。
原文摘要 · Abstract (English)
The integration of Artificial Intelligence (AI) into high-stakes domains such as healthcare, finance, and autonomous systems is often constrained by concerns over transparency, interpretability, and trust. While Human-Centered AI (HCAI) emphasizes alignment with human values, Explainable AI (XAI) enhances transparency by making AI decisions more understandable. However, the lack of a unified approach limits AI's effectiveness in critical decision-making scenarios. This paper presents a novel three-layered framework that bridges HCAI and XAI to establish a structured explainability paradigm. The framework comprises (1) a foundational AI model with built-in explainability mechanisms, (2) a human-centered explanation layer that tailors explanations based on cognitive load and user expertise, and (3) a dynamic feedback loop that refines explanations through real-time user interaction. The framework is evaluated across healthcare, finance, and software development, demonstrating its potential to enhance decision-making, regulatory compliance, and public trust. Our findings advance Human-Centered Explainable AI (HCXAI), fostering AI systems that are transparent, adaptable, and ethically aligned.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。