AIM is a new saliency-guided adversarial feature replacement method to evaluate faithfulness of saliency maps and reliability of masking operators on image, audio, and EEG tasks.
Fidelity of interpretability methods and perturba- tion artifacts in neural networks
2 Pith papers cite this work, alongside 2 external citations. Polarity classification is still indexing.
2
Pith papers citing it
2
external citations · external index
fields
cs.LG 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
A latent mediation framework with sparse autoencoders enables non-additive token-level influence attribution in LLMs by learning orthogonal features and back-propagating attributions.
citing papers explorer
-
AIM: Adversarial Information Masking for Faithfulness Evaluation of Saliency Maps
AIM is a new saliency-guided adversarial feature replacement method to evaluate faithfulness of saliency maps and reliability of masking operators on image, audio, and EEG tasks.
-
Correcting Influence: Unboxing LLM Outputs with Orthogonal Latent Spaces
A latent mediation framework with sparse autoencoders enables non-additive token-level influence attribution in LLMs by learning orthogonal features and back-propagating attributions.