让大模型解释更小更快，用抽象方法自动删减无关特征。
FAME: Formal Abstract Minimal Explanation for Neural Networks
- 用专门设计的扰动域替代遍历顺序，加速解释生成。
- 在中大型网络上实现更小解释且速度更快，比VERIX+提升明显。
- 首次提供可量化解释质量的评估方法，适合可信AI研究者。
我们提出FAME（形式化抽象最小解释），一种基于抽象解释的新型归纳解释方法。FAME是首个能扩展到大规模神经网络的同时缩小解释规模的方法。其核心贡献在于设计专用扰动域，避免依赖遍历顺序。FAME通过逐步收缩这些域，并利用LiRPA边界剔除无关特征，最终收敛至形式化的抽象最小解释。为评估解释质量，我们引入一种程序，通过对抗攻击结合可选的VERIX+精炼步骤，测量抽象最小解释与真实最小解释之间的最坏距离。我们在中到大规模神经网络上对FAME与VERIX+进行基准测试，结果表明其在解释规模和运行时间上均保持一致优势。
原文摘要 · Abstract (English)
We propose FAME (Formal Abstract Minimal Explanations), a new class of abductive explanations grounded in abstract interpretation. FAME is the first method to scale to large neural networks while reducing explanation size. Our main contribution is the design of dedicated perturbation domains that eliminate the need for traversal order. FAME progressively shrinks these domains and leverages LiRPA-based bounds to discard irrelevant features, ultimately converging to a formal abstract minimal explanation. To assess explanation quality, we introduce a procedure that measures the worst-case distance between an abstract minimal explanation and a true minimal explanation. This procedure combines adversarial attacks with an optional VERIX+ refinement step. We benchmark FAME against VERIX+ and demonstrate consistent gains in both explanation size and runtime on medium- to large-scale neural networks.
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。