T$^\text{3}$SVFND: Towards an Evolving Fake News Detector for Emergencies with Test-time Training on Short Video Platforms

TL;DR

This paper introduces T3SVFND, a test-time training framework that improves fake news video detection during emergencies by adapting to distribution shifts using a self-supervised MLM task across multiple modalities.

Contribution

It proposes a novel test-time training approach with a self-supervised MLM task to enhance robustness in fake news video detection, especially in emergency scenarios.

Findings

Significantly improves detection accuracy during emergencies

Effective adaptation to distribution shifts in test data

Outperforms existing methods on benchmark datasets

Abstract

The existing methods for fake news videos detection may not be generalized, because there is a distribution shift between short video news of different events, and the performance of such techniques greatly drops if news records are coming from emergencies. We propose a new fake news videos detection framework (T$^3$SVFND) using Test-Time Training (TTT) to alleviate this limitation, enhancing the robustness of fake news videos detection. Specifically, we design a self-supervised auxiliary task based on Mask Language Modeling (MLM) that masks a certain percentage of words in text and predicts these masked words by combining contextual information from different modalities (audio and video). In the test-time training phase, the model adapts to the distribution of test data through auxiliary tasks. Extensive experiments on the public benchmark demonstrate the effectiveness of the proposed model, especially for the detection of emergency news.