A Symbolic Adversarial Learning Framework for Evolving Fake News Generation and Detection

TL;DR

This paper introduces SALF, a symbolic adversarial learning framework that enhances fake news generation and detection by iterative, language-based adversarial interactions, significantly challenging current detection methods and improving detector robustness.

Contribution

The work presents a novel symbolic adversarial training paradigm for fake news, moving beyond neural updates to language-based agent interactions, improving the generation of sophisticated fake news and detector refinement.

Findings

SALF degrades state-of-the-art detection performance by up to 53.4% in Chinese and 34.2% in English.

SALF improves detection of refined fake news by up to 7.7%.

Experiments on multilingual datasets validate SALF's effectiveness.

Abstract

Rapid LLM advancements heighten fake news risks by enabling the automatic generation of increasingly sophisticated misinformation. Previous detection methods, including fine-tuned small models or LLM-based detectors, often struggle with its dynamically evolving nature. In this work, we propose a novel framework called the Symbolic Adversarial Learning Framework (SALF), which implements an adversarial training paradigm by an agent symbolic learning optimization process, rather than relying on numerical updates. SALF introduces a paradigm where the generation agent crafts deceptive narratives, and the detection agent uses structured debates to identify logical and factual flaws for detection, and they iteratively refine themselves through such adversarial interactions. Unlike traditional neural updates, we represent agents using agent symbolic learning, where learnable weights are defined by agent prompts, and simulate back-propagation and gradient descent by operating on natural language representations of weights, loss, and gradients. Experiments on two multilingual benchmark datasets demonstrate SALF's effectiveness, showing it generates sophisticated fake news that degrades state-of-the-art detection performance by up to 53.4% in Chinese and 34.2% in English on average. SALF also refines detectors, improving detection of refined content by up to 7.7%. We hope our work inspires further exploration into more robust, adaptable fake news detection systems.