Adaptive Equilibrium: Dynamic Weighting Framework for Generalized Interruption of DeepFake Models

TL;DR

This paper introduces the Adaptive Equilibrium Framework (AEF), a dynamic weighting method that improves the uniformity and effectiveness of deepfake model disruption by adaptively balancing interference across diverse architectures.

Contribution

The paper presents a novel adaptive weighting mechanism that addresses the imbalance in deepfake disruption, enabling more consistent performance across different models.

Findings

AEF achieves more balanced interruption success rates.

Dynamic weighting improves robustness against resistant models.

Experiments validate uniform effectiveness across architectures.

Abstract

The advancement of generalized deepfake disruption is constrained by the interruption imbalance, a fundamental bottleneck inherent to the generation of universal perturbations. We reveal that conventional static gradient normalization fundamentally struggles to resolve architectural conflicts, causing the optimization to bias towards susceptible models while neglecting resistant ones. We argue that achieving high and uniform effectiveness requires resolving this imbalance by reaching an adaptive equilibrium. We propose the Adaptive Equilibrium Framework (AEF), which employs a dynamic weighting mechanism that utilizes real-time loss feedback to adaptively assign greater interruption weights to the most resistant models. This approach shifts the optimization from an average-case problem to finding a dynamic balance, driving the perturbation to a uniformly effective equilibrium state. Comprehensive experiments validate that AEF achieves a more balanced interruption performance, maintaining a consistent interruption success rate across the evaluated diverse architectures.