Lei, Y.
, Kacker, R.
and Kuhn, D.
(2025),
ABLE: Using Adversarial Pairs to Construct Local Models for Explaining Model Predictions, 2026 ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Jeju, KR
(Accessed May 21, 2025)
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
Secure .gov websites use HTTPS
A lock (
) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.
ABLE: Using Adversarial Pairs to Construct Local Models for Explaining Model Predictions
Published
Author(s)
Yu Lei, ,
Abstract
Machine learning models are increasingly used in critical applications but are mostly "black boxes" due to their lack of transparency. Local explanation approaches, such as LIME [26], address this issue by approximating the behavior of complex models near a test instance using simple, interpretable models. However, these approaches suffer from instability and poor local fidelity. In this paper, we propose a novel approach called Adversarial Bracketed Local Explanation (ABLE) that overcomes these limitations.Conference Dates
August 3-7, 2026
Conference Location
Jeju, KR
Conference Title
2026 ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
Pub Type
Conferences
Citation
Issues
If you have any questions about this publication or are having problems accessing it, please contact [email protected].
Created March 26, 2025, Updated May 16, 2025