The paper proposes Behaviorally-Adaptive Visual Diversion (BAVD), a theoretical framework that overlays assessment content with a synthetic, non-semantic visual field whose intensity adapts to observed candidate behavior. The underlying content is unchanged; only its visual presentation is modified to reduce the usefulness of unauthorized screenshots or screen sharing. BAVD also includes accessibility-aware attenuation for candidates with approved visual-processing accommodations. The authors model the system with coupled dynamical components, including a Diversion Field Generator, Rendering Tensor, Behavior Tensor, Composite Integrity Functional, and Multi-dimensional Entropy Model. They establish theoretical properties concerning content fidelity, rendering stability, entropy boundedness, integrity tracking, and closed-loop adaptation stability, while identifying deployment assumptions and the need for empirical validation.
No heat snapshots are available in the last 24 hours.