Faithfulness Without Anchors: A Position Paper on the Structural Misalignments of Explainable AI Faithfulness Metrics for Feature Importance

查看原文 / View Original 编辑 / Edit

Computer Science > Artificial Intelligence arXiv:2610.12493 (cs) [Submitted on 24 Sep 2026] Title: Faithfulness Without Anchors: A Position Paper on the Structural Misalignments of Explainable AI Faithfulness Metrics for Feature Importance Authors: Marco Zullich, Emily Schiller, Ivan Gentile, David Dembinsky, Adriano Lucieri, Kilian Göller, Steffen Seitz.

计算机科学 > 人工智能 arXiv:2610.12493 (cs) [提交于 2026 年 9 月 24 日] 标题：无锚点的忠实度：关于可解释人工智能特征重要性忠实度指标结构性失调的立场文件作者：Marco Zullich, Emily Schiller, Ivan Gentile, David Dembinsky, Adriano Lucieri, Kilian Göller, Steffen Seitz。

Abstract: Explainable Artificial Intelligence (XAI) tools have often been presented as methods to increase trust in machine learning models by making aspects of their algorithms more human-understandable. However, evidence accumulated over the last decade points to an overall failure of this objective, instead promoting overreliance on incorrect predictions, which can subsequently lead to algorithmic aversion.

摘要：可解释人工智能（XAI）工具通常被视为通过使算法的某些方面更易于人类理解来增加对机器学习模型信任的方法。然而，过去十年积累的证据表明，这一目标总体上是失败的，反而助长了对错误预测的过度依赖，进而可能导致对算法的厌恶。

Rather, XAI appears to function primarily as a debugging tool, useful for obtaining anecdotal insights into, for example, features learned by the model. One of the main issues underlying this limitation is that XAI methods often oversimplify the inner workings of the model, resulting in misalignment between the explanation and the model itself - a behavior referred to as unfaithfulness or infidelity.

相反，XAI 似乎主要作为一种调试工具发挥作用，有助于获得关于模型所学习特征等方面的轶事见解。导致这一局限性的主要问题之一是，XAI 方法往往过度简化了模型的内部运作，导致解释与模型本身之间出现失调——这种行为被称为“不忠实”或“不诚实”。

Faithfulness can be interpreted as the degree of approximation between an explanation and the model. Computing faithfulness in a rigorous manner would thus make it possible to quantify one of the various elements of trustworthiness of an explanation. However, existing metrics present significant limitations, making this quantification either infeasible or computationally expensive. In this position paper, we outline five major shortcomings of current faithfulness metrics and propose four possible avenues for research to bolster their reliability.

忠实度可以被理解为解释与模型之间的近似程度。因此，以严谨的方式计算忠实度，将使量化解释可信度的各种要素之一成为可能。然而，现有的指标存在显著局限性，使得这种量化要么不可行，要么计算成本高昂。在这篇立场文件中，我们概述了当前忠实度指标的五个主要缺陷，并提出了四个可能的研究方向，以增强其可靠性。