Generating label cohesive and well-formed adversarial claims

P Atanasova - Accountable and Explainable Methods for Complex …, 2024 - Springer
Adversarial attacks reveal important vulnerabilities and flaws of trained models. One potent
type of attack are universal adversarial triggers, which are individual n-grams that, when …

Towards Explainable Fact Checking

I Augenstein - arXiv preprint arXiv:2108.10274, 2021 - arxiv.org
The past decade has seen a substantial rise in the amount of mis-and disinformation online,
from targeted disinformation campaigns to influence politics, to the unintentional spreading …

[图书][B] Accountable and Explainable Methods for Complex Reasoning over Text

P Atanasova - 2024 - Springer
A major concern with machine learning (ML) models is their opacity. They are deployed in
an increasing number of applications where they often operate as black boxes that do not …

System and method for generating adversarial examples

DIA Ousmane - US Patent 11,157,772, 2021 - Google Patents
Methods and systems for generating adversarial examples are disclosed. The method
comprises accessing a set of inputs and generating an instance of a variable auto-encoder …