🤖 AI Summary
To address the trustworthiness bottleneck in intelligent network services—stemming from weak robustness, high security risks, and insufficient fairness of AIGC models—this paper proposes TrustGAIN, an end-to-end trustworthy framework. Methodologically, it establishes the first unified “robustness–security–fairness” tri-dimensional evaluation and assurance framework tailored for network services; innovatively integrates emotion-aware unsafe content detection with adversarial robust training, multimodal content safety filtering, fairness-constrained optimization, and emotion-guided generation strategies. Evaluated on high-risk datasets—including fake news, malicious code, and unsafe reviews—using large language models (LLMs), TrustGAIN achieves significant improvements: +8.2% detection accuracy and +31.5% robustness against adversarial attacks. Its deployment feasibility and effectiveness are further validated across multiple real-world AIGC service scenarios.
📝 Abstract
AI-generated content (AIGC) models, represented by large language models (LLM), have revolutionized content creation. High-speed next-generation communication technology is an ideal platform for providing powerful AIGC network services. At the same time, advanced AIGC techniques can also make future network services more intelligent, especially various online content generation services. However, the significant untrustworthiness concerns of current AIGC models, such as robustness, security, and fairness, greatly affect the credibility of intelligent network services, especially in ensuring secure AIGC services. This paper proposes TrustGAIN, a trustworthy AIGC framework that incorporates robust, secure, and fair network services. We first discuss the robustness to adversarial attacks faced by AIGC models in network systems and the corresponding protection issues. Subsequently, we emphasize the importance of avoiding unsafe and illegal services and ensuring the fairness of the AIGC network services. Then as a case study, we propose a novel sentiment analysis-based detection method to guide the robust detection of unsafe content in network services. We conduct our experiments on fake news, malicious code, and unsafe review datasets to represent LLM application scenarios. Our results indicate that TrustGAIN is an exploration of future networks that can support trustworthy AIGC network services.