Attacks against Abstractive Text Summarization Models through Lead Bias and Influence Functions

TL;DR

This work unveils a novel approach by exploiting the inherent lead bias in summarization models, to perform adversarial perturbations and introduces an innovative application of influence functions, to execute data poisoning, which compromises the model's integrity.

Abstract

Large Language Models have introduced novel opportunities for text comprehension and generation. Yet, they are vulnerable to adversarial perturbations and data poisoning attacks, particularly in tasks like text classification and translation. However, the adversarial robustness of abstractive text summarization models remains less explored. In this work, we unveil a novel approach by exploiting the inherent lead bias in summarization models, to perform adversarial perturbations. Furthermore, we introduce an innovative application of influence functions, to execute data poisoning, which compromises the model's integrity. This approach not only shows a skew in the models behavior to produce desired outcomes but also shows a new behavioral change, where models under attack tend to generate extractive summaries rather than abstractive summaries.

Figures (3)

• Figure 1: Framework showing implementation of adversarial perturbations
• Figure 2: Poisoning attack using Influence Functions
• Figure 3: Results demonstrating the percentage of summaries exhibiting behavioral shift after data poisoning