让AI决策与企业目标对齐，支持多角色实时干预和审计。
HADA: Human-AI Agent Decision Alignment Architecture
- 用角色代理包装各类算法，实现人机协同决策控制
- 可追踪所有决策的指标、约束与完整溯源链路
- 适合金融、医疗等需高透明度与合规性的场景
我们提出HADA（Human-AI Agent Decision Alignment），一种与协议和框架无关的参考架构，确保大型语言模型（LLM）代理与传统算法始终与组织目标和价值观保持一致。HADA通过为业务、数据科学、审计、伦理和客户等角色创建专用代理，每个代理提供对话式API，使技术与非技术人员均可在战略、战术和实时层面查询、引导、审计或质疑每项决策。对齐目标、关键绩效指标（KPI）和价值约束以自然语言表达，并持续传播、记录和版本化，即使在不同编排栈上运行数千个异构代理也能保持一致。一个基于云原生的演示原型部署了生产级信贷评分模型（getLoanDecision），使用Docker/Kubernetes/Python；五个脚本化的零售银行场景展示了目标变更、参数调整、解释请求及伦理触发器在全链路中的端到端流转。评估遵循设计科学研究方法，观察与日志检查表明完全覆盖六个预设目标：每个角色均可发起对话控制，追溯KPI与价值约束，检测并缓解基于邮政编码的偏见，并复现完整决策溯源链，且不依赖底层LLM或代理库。贡献包括：（1）开源的HADA架构，（2）多智能体系统中人机对齐的中等规模设计理论，（3）实证证据证明框架无关、协议合规的角色代理能提升现实决策流水线的准确性、透明度与合规性。
原文摘要 · Abstract (English)
We present HADA (Human-AI Agent Decision Alignment), a protocol- and framework agnostic reference architecture that keeps both large language model (LLM) agents and legacy algorithms aligned with organizational targets and values. HADA wraps any algorithm or LLM in role-specific stakeholder agents -- business, data-science, audit, ethics, and customer -- each exposing conversational APIs so that technical and non-technical actors can query, steer, audit, or contest every decision across strategic, tactical, and real-time horizons. Alignment objectives, KPIs, and value constraints are expressed in natural language and are continuously propagated, logged, and versioned while thousands of heterogeneous agents run on different orchestration stacks. A cloud-native proof of concept packages a production credit-scoring model (getLoanDecision) and deploys it on Docker/Kubernetes/Python; five scripted retail-bank scenarios show how target changes, parameter tweaks, explanation requests, and ethics triggers flow end to end through the architecture. Evaluation followed the Design-Science Research Methodology. Walkthrough observation and log inspection demonstrated complete coverage of six predefined objectives: every role could invoke conversational control, trace KPIs and value constraints, detect and mitigate ZIP-code bias, and reproduce full decision lineage, independent of the underlying LLM or agent library. Contributions: (1) an open-source HADA architecture, (2) a mid-range design theory for human-AI alignment in multi-agent systems, and (3) empirical evidence that framework-agnostic, protocol-compliant stakeholder agents improve accuracy, transparency, and ethical compliance in real-world decision pipelines.
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。