analiza aiqbench
This research introduces TNC-Defense, a gray-box framework that detects and mitigates backdoor attacks in diffusion models by exploiting temporal noise consistency anomalies without requiring access to model weights. The approach significantly improves detection accuracy while effectively suppressing malicious triggers with minimal impact on generation quality, addressing critical security gaps in deployed AIGC services.
Backdoor Sentinel: Detecting and Detoxifying Backdoors in Diffusion Models via Temporal Noise Consistency
arXiv:2602.01765v2 Announce Type: replace-cross Abstract: Diffusion models have been widely deployed in AIGC services, but their reliance on opaque training data exposes them to backdoor attacks. In practical auditing scenarios, auditors are typically unable to access model parameters due to intellectual property protection, making white-box or query-intensive detection impractical. After detection, existing detoxification approaches are trapped in a dilemma between detoxification effectiveness and generation quality for service providers. We reveal Temporal Noise Consistency (TNC), a previously unreported phenomenon in which backdoor activation disrupts the consistency of noise predictions between adjacent diffusion timesteps within specific temporal segments, while clean inputs remain stable. Based on this finding, we propose TNC-Defense, a closed-loop framework for gray-box backdoor detection and model repair. Specifically, TNC-Detect (for auditors) uses inference-stage adjacent-noise statistics to detect backdoors and precisely localize anomalous timesteps without model-weight access. TNC-Detox (for service providers) utilizes these locations to perform trigger-agnostic, timestep-aware correction of the generation path, suppressing backdoor behavior while reducing detoxification cost. Across five representative backdoor attacks and state-of-the-art defenses, TNC-Defense improves the average detection accuracy by $11\%$ with negligible additional overhead, and invalidates an average of $98.5\%$ of triggered samples with only a mild degradation in generation quality. Our code is publicly available at: https://github.com/binzhwang/TNC-Defense.
Abstract:Diffusion models have been widely deployed in AIGC services, but their reliance on opaque training data exposes them to backdoor attacks. In practical auditing scenarios, auditors are typically unable to access model parameters due to intellectual property protection, making white-box or query-intensive detection impractical. After detection, existing detoxification approaches are trapped in a dilemma between detoxification effectiveness and generation quality for service providers. We reveal Temporal Noise Consistency (TNC), a previously unreported phenomenon in which backdoor activation disrupts the consistency of noise predictions between adjacent diffusion timesteps within specific temporal segments, while clean inputs remain stable. Based on this finding, we propose TNC-Defense, a closed-loop framework for gray-box backdoor detection and model repair. Specifically, TNC-Detect (for auditors) uses inference-stage adjacent-noise statistics to detect backdoors and precisely localize anomalous timesteps without model-weight access. TNC-Detox (for service providers) utilizes these locations to perform trigger-agnostic, timestep-aware correction of the generation path, suppressing backdoor behavior while reducing detoxification cost. Across five representative backdoor attacks and state-of-the-art defenses, TNC-Defense improves the average detection accuracy by $11\%$ with negligible additional overhead, and invalidates an average of $98.5\%$ of triggered samples with only a mild degradation in generation quality. Our code is publicly available at: this https URL.
Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
arXiv:2602.01765 [cs.CR]
(or arXiv:2602.01765v2 [cs.CR] for this version)
https://doi.org/10.48550/arXiv.2602.01765 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 297 cuvinte
Subiecte conexe
Frontier Risk Report (February to March 2026)
METR · 19 mai 2026
Frontier AI Risk Report (February–March 2026)
METR · 19 mai 2026
Informe de riesgos de la IA de frontera (febrero–marzo de 2026)
METR · 19 mai 2026
Learning Normal Diffusion Dynamics for Backdoor Defense in Text-to-Image Models
arXiv · 1 oct. 2026