Incorruptible Neural Networks: Training Models that can Generalize to Large Internal Perturbations

Philip Jacobson, Ben Feinberg, Suhas Kumar, Sapan Agarwal, T. Patrick Xiao, Christopher H. Bennett

BibTeX
@misc{jacobson2026incorruptible,
  author = {Philip Jacobson and Ben Feinberg and Suhas Kumar and Sapan Agarwal and T. Patrick Xiao and Christopher H. Bennett},
  title = {{Incorruptible Neural Networks: Training Models that can Generalize to Large Internal Perturbations}},
  year = {2026},
  eprint = {2602.07320},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2602.07320},
}

Loading related content...

Figure