EvoGuard: An Extensible Agentic RL-based Framework for Practical and Evolving AI-Generated Image Detection

TL;DR

EvoGuard is an agentic framework that intelligently orchestrates multiple detectors for AI-generated image detection, achieving state-of-the-art accuracy while being adaptable, cost-effective, and capable of integrating new tools without retraining.

Contribution

It introduces a novel agentic, tool-coordinating framework for AIGI detection that leverages reinforcement learning and reflection, surpassing traditional methods in accuracy and extensibility.

Findings

Achieves state-of-the-art detection accuracy.

Effectively integrates heterogeneous detectors.

Reduces reliance on costly annotations.

Abstract

The rapid proliferation of AI-Generated Images (AIGIs) has introduced severe risks of misinformation, making AIGI detection a critical yet challenging task. While traditional detection paradigms mainly rely on low-level features, recent research increasingly focuses on leveraging the general understanding ability of Multimodal Large Language Models (MLLMs) to achieve better generalization, but still suffer from limited extensibility and expensive training data annotations. To better address complex and dynamic real-world environments, we propose EvoGuard, a novel agentic framework for AIGI detection. It encapsulates various state-of-the-art (SOTA) off-the-shelf MLLM and non-MLLM detectors as callable tools, and coordinates them through a capability-aware dynamic orchestration mechanism. Empowered by the agent's capacities for autonomous planning and reflection, it intelligently selects suitable tools for given samples, reflects intermediate results, and decides the next action, reaching a final conclusion through multi-turn invocation and reasoning. This design effectively exploits the complementary strengths among heterogeneous detectors, transcending the limits of any single model. Furthermore, optimized by a GRPO-based Agentic Reinforcement Learning algorithm using only low-cost binary labels, it eliminates the reliance on fine-grained annotations. Extensive experiments demonstrate that EvoGuard achieves SOTA accuracy while mitigating the bias between positive and negative samples. More importantly, it allows the plug-and-play integration of new detectors to boost overall performance in a train-free manner, offering a highly practical, long-term solution to ever-evolving AIGI threats. Source code will be publicly available upon acceptance.