This work introduces a model-agnostic and explanation-agnostic index for quantifying the temporal consistency of automated explanations, and empirically demonstrates that the proposed index facilitates the quantification and localization of temporal instability in explanation streams.
A framework for the evaluation of XAI methods based on synthetic ground truth that relies on controlled interventions to generate synthetic datasets in which the importance of input components can be determined by design, allowing a comprehensive assessment of explanation methods in heterogeneous settings.
Miquel Miró-Nicolau, F. Spinnato, Riccardo Guidotti· 0 citations
The limitations of Explainable Artificial Intelligence (XAI) with respect to insufficient evaluation are illustrated through the DetoxAI image recognition system for bias detection and concept unlearning and methods for adapting explanations to evolving data streams with concept drift are explored.
It is demonstrated empirically that ProToMEx not only produces explanations of comparable fidelity to popular methods like SHAP and LIME but also drastically reduces the amortised computational cost of generating local explanations, making it highly suitable for real-time applications.
A. Georgara, Adarsh Valoor, Sarvapali D. Ramchurn· 0 citations
This tutorial provides a comprehensive, end-to-end view of LLM interpretability, transitioning from microscopic neural analysis to macroscopic application and deployment, and explores how these interpretability paradigms scale and inspire the design of frontier architectures, agentic systems, and thinking models.
Wei Zhang, Zheng-Fu He, Lu-Lu Zhang et al.· Proceedings of the 32nd ACM...· 0 citations
Explanations of machine learning models are usually judged by criteria that are hard to compare. We propose a simpler test: if an explanation really describes how a model uses its features, it should be possible to rebuild the model's predictions from it. We turn each explanation into a predictor by reading each featur...
This work develops a supervised-compression initializer that provides a data-driven initial guess for the number of clusters and proposes likelihood-based procedures for confidence intervals and a fixed-gate bootstrap method for prediction intervals, demonstrating that MoE models can serve as a flexible, interpretable,...
Jyun-Yu Chen, Ming-Chung Chang, Min Yang et al.· Journal of Statistical Theor...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
