用自然语言查询视觉模型的全局解释，让普通人也能轻松理解黑箱模型决策。
GLARE: A Natural Language Interface for Querying Global Explanations
- 用大模型将自然语言转为结构化查询，自动分析解释数据。
- 在多个数据集上准确响应新问题，对语言错误也具备鲁棒性。
- 适合想快速理解模型决策逻辑的研究者和开发者使用。
尽管全局解释对理解视觉模型在不同数据集、类别和决策情境下的行为至关重要，但其复杂且静态的特性常阻碍实际探索。用户通常希望针对特定问题获得精准回答，而非被动接收固定解释。为此，我们提出一种基于大模型的交互式界面（GLARE），支持通过自然语言查询黑箱图像分类器的全局解释。系统核心大模型作为中介，将自然语言问题转化为对局部解释数据的结构化SQL查询，实现灵活聚合而不暴露底层表示。每个查询返回包含统计信息的自然语言响应，支持局部解释与意图对齐的可视化。我们在意图识别、查询映射准确率、新问题与新数据集泛化能力以及语言错误鲁棒性方面进行评估。结果表明，大模型中介查询显著提升了全局解释的人类可访问性与可用性。
原文摘要 · Abstract (English)
While global explanations are crucial for understanding vision models across datasets, classes, and decision contexts, their complex and monolithic nature often hinders practical exploration. Because users typically seek targeted answers to specific questions rather than static artifacts, we present an LLM-based interactive interface that provides natural language access to global explanations for black-box image classifiers. The system's core LLM acts as a mediator, translating natural language questions into structured SQL queries over local explanation data. This enables flexible aggregation without exposing users to low-level representations. For each query, the interface outputs statistics-augmented natural language responses, supporting local explanations, and intent-aligned visualizations. We evaluate the system on intent interpretation, query mapping accuracy, generalization to novel queries and datasets, and robustness to linguistic errors. Our results demonstrate that LLM-mediated querying substantially improves the accessibility and usability of global explanations for human-centered XAI.
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。