The evidence behind the guides

Sources

Find a study, check what it supports, and follow it back to a concern.

TopicCollection

Sources for Post-hoc explanation unreliability

2 sources

Guide citations are still under editorial review.

Slack and colleagues · 2020 Research study

Fooling LIME and SHAP: Adversarial Attacks on Post hoc Explanation Methods (original source, new tab)

Demonstrates manipulated explanations in tested perturbation-based methods and settings.

Adebayo and colleagues · NeurIPS 2018 Research study

Sanity Checks for Saliency Maps (original source, new tab)

Tests whether selected visual explanation methods depend on the model and training data; not a rejection of all interpretability methods.

Sources retain their original publishers’ rights. Inclusion does not imply endorsement. Saved sources stay in this browser on this device. Dates and source descriptions come from the existing editorial records.