Comparables XAI: Faithful Example-based AI Explanations with Counterfactual Trace Adjustments
This work proposes Comparables XAI for relatable example-based explanations of AI with Trace adjustments that trace counterfactual changes from each Comparable to the Subject, one attribute at a time, monotonically along the AI feature space.
Yi-Fan Zhang, Tianle Ren, Fei Wang et al.
· International Conference on... · 1 citation