AI 中文总结
针对现有XAI叙事无法适配不同受众的问题，提出XstrAI多智能体框架，结合三类LLM智能体与修正循环，在糖尿病、中风风险预测任务中，其叙事适配性与保真度优于多数基线。
AI 中文摘要
SHAP等特征归因方法能为单个模型预测提供有用证据，但其数值输出往往无法满足具备不同专业知识、目标及误释风险的受众需求。在医疗AI领域，同一份局部解释需以截然不同的沟通形式传递给患者、临床医生和数据科学家，而通过大语言模型（LLMs）进行的朴素语言表述易出现依据薄弱、将归因与因果语言混淆、输出有说服力却不符合底层模型证据等问题。我们提出XstrAI，一种受众感知的多智能体框架，该框架将局部解释视为固定证据，并对向每个目标读者的传递方式进行结构化处理。每个预测案例被编码为不可变的结构化表示，在不同受众间保持一致，从而确保底层证据固定。生成过程被分解为三个专门的LLM智能体，分别负责受众感知规划、语言实现以及依据、归因一致性、沟通风险和受众适配性的验证，若检测到不一致则触发有界修正循环。我们在糖尿病和中风风险预测任务上，将XstrAI与11个基线方法（从直接语言表述到重新实现的最先进叙事器）进行了评估。评估结合了两种机制：叙事内机制，用于衡量对SHAP证据的保真度；叙事外机制，通过参考语料库、多家族LLM评判器和目标读者调查来评估受众适配性。在两项评估中，独立评判者均能将XstrAI的叙事准确分配给其 intended audience，且在临床医生和患者受众上优于所有基线方法，在数据科学家受众上表现具有竞争力，而针对受众的单提示基线方法在该受众上表现更优。
英文摘要
Feature-attribution methods such as SHAP provide useful evidence about individual model predictions, but their numerical outputs are rarely sufficient for audiences with different expertise, goals, and risks of misinterpretation. In medical AI, the same local explanation must reach patients, clinicians, and data scientists through markedly different forms of communication, and naive verbalization through large language models (LLMs) is prone to weak grounding, conflation of attribution with causal language, and outputs that are persuasive without being faithful to the underlying model evidence. We introduce XstrAI, an audience-aware multi-agent framework that treats local explanations as fixed evidence and structures how it is communicated to each target reader. Each prediction case is encoded as an immutable structured representation, shared identically across audiences so the underlying evidence remains fixed. Generation is factored into three specialized LLM agents responsible for audience-aware planning, linguistic realization, and validation for grounding, attribution consistency, communicative risk, and audience appropriateness, with a bounded revision loop triggered on detected inconsistencies. We evaluate XstrAI on diabetes and stroke risk prediction against 11 baselines, ranging from direct verbalization to a re-implementation of a state-of-the-art narrator. The evaluation combines an intra-narrative regime measuring fidelity to SHAP evidence with an extra-narrative regime assessing audience appropriateness through reference corpora, multi-family LLM judges, and a survey with target readers. In both evaluations, XstrAI's narratives are consistently assigned to their intended audience by independent judges, and preferred over all baselines on Clinician and Patient audiences, with competitive performance on Data Scientist, where audience-conditioned single-prompt baselines lead.
