Fake News Court: A Multi-Agent Adversarial Framework for Robust Detection of LLM-Generated Fake News

ACL ARR 2026 January Submission9402 Authors

Keywords: Fake News Detection, Muti-Agent, Adversarial Learning, Large Language model

Abstract: Large language models (LLMs) have achieved remarkable success across many domains, yet their generative capabilities have been misused to produce highly realistic fake news that threatens social stability. However, the main existing detection methods, including small language model (SLM)-based approaches and LLM-based detectors, exhibit substantial performance degradation on LLM-generated content, reflecting a lack of robustness. SLM-based detectors rely heavily on fixed data distributions and shallow textual cues, while LLM-based detectors depend on prompt-based inference, making them sensitive to prompting and vulnerable to LLM hallucinations. To address this challenge, we propose Fake News Court (FNC), a robust multi-agent adversarial framework that integrates the complementary strengths of LLM and SLM for detecting LLM-generated fake news. FNC adopts an adversarial learning paradigm in which a quality-controlled generator produces diverse and challenging fake news, while a hybrid detector combines multi-dimensional LLM-based agent reasoning with an SLM-based classifier to ensure stable decisions. Extensive experiments show that FNC improves detection accuracy by an average of 6\% on LLM-generated fake news than the state-of-the-art method, confirming its robustness.

Paper Type: Long

Research Area: AI/LLM Agents

Research Area Keywords: LLM agents, multi-agent systems

Languages Studied: English

Submission Number: 9402