Evaluating Deception Detection Model Robustness To Linguistic Variation

TL;DR

This paper analyzes the robustness of deception detection models to linguistic variations, revealing that character-based defenses are most effective against adversarial attacks in misinformation detection.

Contribution

It provides an extensive evaluation of model robustness to linguistic manipulation, comparing multiple embeddings and defense strategies in deception detection.

Findings

Character or mixed ensemble models are most effective defenses.

Character perturbation attacks are more successful.

Models show high confidence misclassifications and failure modes.

Abstract

With the increasing use of machine-learning driven algorithmic judgements, it is critical to develop models that are robust to evolving or manipulated inputs. We propose an extensive analysis of model robustness against linguistic variation in the setting of deceptive news detection, an important task in the context of misinformation spread online. We consider two prediction tasks and compare three state-of-the-art embeddings to highlight consistent trends in model performance, high confidence misclassifications, and high impact failures. By measuring the effectiveness of adversarial defense strategies and evaluating model susceptibility to adversarial attacks using character- and word-perturbed text, we find that character or mixed ensemble models are the most effective defenses and that character perturbation-based attack tactics are more successful.