Search Results for author: Vansh Gupta

Found 2 papers, 1 papers with code

BITE: Textual Backdoor Attacks with Iterative Trigger Injection

1 code implementation25 May 2022 Jun Yan, Vansh Gupta, Xiang Ren

We propose BITE, a backdoor attack that poisons the training data to establish strong correlations between the target label and a set of "trigger words".

Backdoor Attack Hate Speech Detection +3

NLPIITR at SemEval-2021 Task 6: RoBERTa Model with Data Augmentation for Persuasion Techniques Detection

no code implementations SEMEVAL 2021 Vansh Gupta, Raksha Sharma

This paper describes and examines different systems to address Task 6 of SemEval-2021: Detection of Persuasion Techniques In Texts And Images, Subtask 1.

Data Augmentation

Cannot find the paper you are looking for? You can Submit a new open access paper.