English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

LIME has emerged as one of the most commonly referenced tools in explainableAI (XAI) frameworks that is integrated into critical machine learningapplications--e.g., healthcare and finance. However, its stability remainslittle explored, especially in the context of text data, due to the uniquetext-space constraints. To address these challenges, in this paper, we firstevaluate the inherent instability of LIME on text data to establish a baseline,and then propose a novel algorithm XAIFooler to perturb text inputs andmanipulate explanations that casts investigation on the stability of LIME as atext perturbation optimization problem. XAIFooler conforms to the constraintsto preserve text semantics and original prediction with small perturbations,and introduces Rank-biased Overlap (RBO) as a key part to guide theoptimization of XAIFooler that satisfies all the requirements for explanationsimilarity measure. Extensive experiments on real-world text datasetsdemonstrate that XAIFooler significantly outperforms all baselines by largemargins in its ability to manipulate LIME's explanations with high semanticpreservability.

Are Your Explanations Reliable? Investigating the Stability of LIME in Explaining Text Classifiers by Marrying XAI and Adversarial Attack