理解大型语言模型行为：通过交互式反事实生成与分析
Understanding Large Language Model Behaviors through Interactive Counterfactual Generation and Analysis
AI总结：
本文提出LLM Analyzer交互式可视化系统，通过反事实分析高效探索LLM行为，生成语义反事实并计算特征归因，经用户研究验证其有效性与人机协作价值。
摘要：
理解大型语言模型（LLM）的行为对于确保其安全可靠的使用至关重要。然而，现有的用于LLM的可解释人工智能（XAI）方法主要依赖于词级解释，这些解释通常计算效率低下，并且与人类推理过程不一致。此外，这些方法往往将解释视为一次性输出，忽视了其固有的交互性和迭代性。在本文中，我们提出了LLM Analyzer，一个交互式可视化系统，通过反事实分析实现对LLM行为的直观高效探索，从而解决上述局限性。我们的系统采用了一种新颖的算法，该算法通过用户定义的粒度级别上的定向移除和替换操作，生成流畅且语义上有意义的反事实。这些反事实被用于计算特征归因分数，然后与具体示例集成在基于表格的可视化中，支持对模型行为的动态分析。一项针对LLM从业者的用户研究和专家访谈证明了该系统的可用性和有效性，强调了在解释过程中让人类作为主动参与者而非被动接收者参与的重要性。
英文摘要：
Understanding the behavior of large language models (LLMs) is crucial for ensuring their safe and reliable use. However, existing explainable AI (XAI) methods for LLMs primarily rely on word-level explanations, which are often computationally inefficient and misaligned with human reasoning processes. Moreover, these methods often treat explanation as a one-time output, overlooking its inherently interactive and iterative nature. In this paper, we present LLM Analyzer, an interactive visualization system that addresses these limitations by enabling intuitive and efficient exploration of LLM behaviors through counterfactual analysis. Our system features a novel algorithm that generates fluent and semantically meaningful counterfactuals via targeted removal and replacement operations at user-defined levels of granularity. These counterfactuals are used to compute feature attribution scores, which are then integrated with concrete examples in a table-based visualization, supporting dynamic analysis of model behavior. A user study with LLM practitioners and interviews with experts demonstrate the system's usability and effectiveness, emphasizing the importance of involving humans in the explanation process as active participants rather than passive recipients.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
