_id,doi,title
11366,10.48550/arXiv.2204.07373,Revisiting the adversarial robustness-accuracy tradeoff in robot learning