← Search

Loris D’Antoni

2 accepted papers

2021

Certified Robustness to Programmable Transformations in LSTMs

EMNLP 2021main

Deep neural networks for natural language processing are fragile in the face of adversarial examples—small input perturbations, like synonym substitution or word duplication, which cause a neural network to change its prediction. We present an approach to certifying the robustness of LSTMs (and exte…

2020

Robustness to Programmable String Transformations via Augmented Abstract Training

ICML 2020poster

Deep neural networks for natural language processing tasks are vulnerable to adversarial input perturbations. In this paper, we present a versatile language for programmatically specifying string transformations—e.g., insertions, deletions, substitutions, swaps, etc.—that are relevant to the task at…