[论文解读] Transferable XAI: Relating Understanding Across Domains with Explanation Transfer

Fei Wang, Yifan Zhang+1 more|arXiv (Cornell University)|Feb 14, 2026

Explainable Artificial Intelligence (XAI)被引用 0

一句话总结

论文提出可迁移的 XAI（Transferable XAI），一种仿射变换框架，用于在相关的 AI 领域之间（子空间、任务和属性）转移解释，并在健康风险与空气污染情境下进行任务和属性转移的评估。

ABSTRACT

Current Explainable AI (XAI) focuses on explaining a single application, but when encountering related applications, users may rely on their prior understanding from previous explanations. This leads to either overgeneralization and AI overreliance, or burdensome independent memorization. Indeed, related decision tasks can share explanatory factors, but with some notable differences; e.g., body mass index (BMI) affects the risks for heart disease and diabetes at the same rate, but chest pain is more indicative of heart disease. Similarly, models using different attributes for the same task still share signals; e.g., temperature and pressure affect air pollution but in opposite directions due to the ideal gas law. Leveraging transfer of learning, we propose Transferable XAI to enable users to transfer understanding across related domains by explaining the relationship between domain explanations using a general affine transformation framework applied to linear factor explanations. The framework supports explanation transfer across various domain types: translation for data subspace (subsuming prior work on Incremental XAI), scaling for decision task, and mapping for attributes. Focusing on task and attributes domain types, in formative and summative user studies, we investigated how well participants could understand AI decisions from one domain to another. Compared to single-domain and domain-independent explanations, Transferable XAI was the most helpful for understanding the second domain, leading to the best decision faithfulness, factor recall, and ability to relate explanations between domains. This framework contributes to improving the reusability of explanations across related AI applications by explaining factor relationships between subspaces, tasks, and attributes.

研究动机与目标

• 帮助用户将一个 AI 领域的理解应用到相关领域，而不过度泛化或记忆负担。
• 引入一个可解释的仿射变换框架，用于子空间、任务和属性之间的解释转移。
• 通过转移映射的稀疏性降低认知负荷，同时保持解释的可信性。
• 在接近真实世界的健康风险与空气污染情景中，对任务转移和属性转移进行形成性与总结性评估。

提出的方法

• 用权重 w 和相对属性值 χ 解释预测的线性因子。
• 定义仿射变换 w_T = A w_O + b，将原域解释映射到目标域解释。
• 实现域类型特定转移：通过平移实现子空间转移（A = I, b ≠ 0）；通过缩放实现任务转移（A = diag(κ); b = 0）；通过映射实现属性转移（w_T = Mχ^T w_O，且稀疏的 Mχ）。
• 通过带有稀疏正则化的原始解释器和目标解释器的联合训练，在保持可信性的同时降低认知负荷（L = L_O + L_T + λ L_s）。
• 在健康风险与空气污染两项实验中评估任务转移与属性转移，以评估可信性、因子回忆和跨域关系理解。

Figure 1. Concept of Transferable XAI for cross-domain understanding. a) Original Domain: Initial understanding of AI decisions. b) Target Domain: Understanding goal for the second domain. c) Mismatched Domains: Applying original understanding to the target domain results in knowledge gaps (gray) an

实验结果

研究问题

• RQ1 研究用户在将原始域的 AI 决策理解通过可迁移的 XAI 转移到目标域的能力有多强？
• RQ2 与单域解释相比，可迁移的 XAI 是否提高了目标域的决策可信性、因子回忆以及跨域解释关系的能力？
• RQ3 子空间、任务与属性转移在认知负荷与有效性方面有何差异？
• RQ4 翻译、缩放与映射转移是否支持面向公众的跨域直观、稀疏解释？

主要发现

• 可迁移的 XAI 在帮助用户理解目标域的 AI 决策方面优于其他解释方法。
• 在目标域中实现了最佳的决策可信性与因子回忆。
• 提升了用户跨域关联解释的能力。
• 稀疏正则化通过限制转移因子数量来降低认知负荷。
• 该框架通过将子空间转移与额外域转移统一起来，包含了增量 XAI 的思想。

Figure 2. Conceptual example of affine transformations across different domain types, shown with 2-dimensional data (two attributes) for simplicity. See technical details in Section 3.3 .