arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2503.05050cs.CLcs.AIcs.LG
收藏

一种用于评估大语言模型中XAI技术有效性的统一框架与新度量指标

A Unified Framework with Novel Metrics for Evaluating the Effectiveness of XAI Techniques in LLMs

College of Engineering and Applied Science, University of Colorado Colorado Springs(科罗拉多大学科罗拉多斯普林斯分校工程与应用科学学院)

Melkamu Abay Mersha, Mesay Gemeda Yigezu, Hassan Shakil, Ali K. AlShami, Sanghyun Byun, Jugal Kalita

AI总结:

本文提出一个含四个新度量指标的统一评估框架,在五个大语言模型和两个任务上评估五种XAI技术,发现LIME综合表现最佳,AMV鲁棒性与一致性突出,LRP对比性优异。

摘要:

大语言模型(LLMs)日益增长的复杂性对其透明度和可解释性提出了重大挑战,因此需要使用可解释人工智能(XAI)技术来增强可信性和可用性。本研究引入了一个包含四个新度量指标的综合评估框架,用于评估五种XAI技术在五个大语言模型和两个下游任务上的有效性。我们应用该框架,使用IMDB电影评论数据集和Tweet Sentiment Extraction数据集,评估了LIME、SHAP、Integrated Gradients、层级相关性传播(LRP)和注意力机制可视化(AMV)等几种XAI技术。评估聚焦于四个关键指标:人类推理一致性(HA)、鲁棒性、一致性和对比性。结果表明,LIME在多个大语言模型和评估指标上持续获得高分,而AMV表现出优越的鲁棒性和近乎完美的一致性。LRP在对比性方面表现突出,尤其是在更复杂的模型上。我们的研究结果为不同XAI方法的优势和局限性提供了有价值的见解,为开发和选择适用于大语言模型的XAI技术提供了指导。

英文摘要:

The increasing complexity of LLMs presents significant challenges to their transparency and interpretability, necessitating the use of eXplainable AI (XAI) techniques to enhance trustworthiness and usability. This study introduces a comprehensive evaluation framework with four novel metrics for assessing the effectiveness of five XAI techniques across five LLMs and two downstream tasks. We apply this framework to evaluate several XAI techniques LIME, SHAP, Integrated Gradients, Layer-wise Relevance Propagation (LRP), and Attention Mechanism Visualization (AMV) using the IMDB Movie Reviews and Tweet Sentiment Extraction datasets. The evaluation focuses on four key metrics: Human-reasoning Agreement (HA), Robustness, Consistency, and Contrastivity. Our results show that LIME consistently achieves high scores across multiple LLMs and evaluation metrics, while AMV demonstrates superior Robustness and near-perfect Consistency. LRP excels in Contrastivity, particularly with more complex models. Our findings provide valuable insights into the strengths and limitations of different XAI methods, offering guidance for developing and selecting appropriate XAI techniques for LLMs.

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。

↑