SOURCE-LINKED INTELLIGENCE

When Temporal Perturbations Act Like Sensor Biases: Label-Free Auditing of Wearable Activity Recognizers

arXiv · AI, language, vision and robotics · article · Sep 24, 2026 · UTC

Wearable human-activity recognition (HAR) models operate across sensors, subjects, and backbones, yet a smooth waveform may appear temporal while exploiting a persistent sensor offset primarily. We introduce SpectrumAudit, a label-sealed audit that fits a phase-randomized full-window stimulus on calibration windows from subjects held out from training and testing. After selection, it replays its exact DC projection and budget-constrained zero-mean residual on the same frozen victim without refitting. Across 27 victims from three datasets and three backbones, the selected waveforms cause 2.87-4

Read original source ↗ Open in workspace

recordType

paper

region

Global