Agent4FaceForgery: Multi-Agent LLM Framework for Realistic Face Forgery Detection
This work addresses the problem of improving face forgery detection for real-world applications like social media security, though it is incremental as it builds on existing simulation and LLM methods.
The paper tackles the gap between offline benchmarks and real-world efficacy in face forgery detection by introducing Agent4FaceForgery, a multi-agent LLM framework that simulates human forgery creation to generate realistic training data, resulting in significant performance gains for detectors across multiple architectures.
Face forgery detection faces a critical challenge: a persistent gap between offline benchmarks and real-world efficacy,which we attribute to the ecological invalidity of training data.This work introduces Agent4FaceForgery to address two fundamental problems: (1) how to capture the diverse intents and iterative processes of human forgery creation, and (2) how to model the complex, often adversarial, text-image interactions that accompany forgeries in social media. To solve this,we propose a multi-agent framework where LLM-poweredagents, equipped with profile and memory modules, simulate the forgery creation process. Crucially, these agents interact in a simulated social environment to generate samples labeled for nuanced text-image consistency, moving beyond simple binary classification. An Adaptive Rejection Sampling (ARS) mechanism ensures data quality and diversity. Extensive experiments validate that the data generated by our simulationdriven approach brings significant performance gains to detectors of multiple architectures, fully demonstrating the effectiveness and value of our framework.