CASPER et al 2022
Robust feature-level adversaries are interpretability tools
neurips 2022
Casper et al Robust feature-level adversaries are interpretability tools
Work to cite in connection with these data:
Casper S, Nadeau M, Hadfield-Mennel D, Kreiman G. Robust feature-level adversaries are interpretability toolsthe human frontal cortex. Neural Information Processing Systems (NeurIPS 2022). PDF
Download data: