Detecting Diffusion-Generated Time Series Under Generator Shift
For researchers and practitioners in time series analysis and generative model detection, this work highlights that image-domain detection methods do not transfer directly to time series, establishing a baseline for a new problem.
The paper explores detection of diffusion-generated time series under unknown generator shift, finding that black-box detection using a simple classifier outperforms white-box reconstruction-based methods, achieving 79.2 average F1 and 57.2 TPR@1%FPR, a 22.1% relative improvement.
The boundary between real and diffusion-generated time series is becoming increasingly difficult to draw, yet detection in this domain remains underexplored, especially when the generator is unknown. We compare white-box detection, which requires access to the generator, against black-box detection, which operates on the raw signal alone. The white-box approach, a reconstruction-based detector adapted from the image domain, works well in in-distribution but breaks down under generator shift: reconstruction-based detection in images succeeds because large generic generators provide a near-universal reconstruction prior, and no analogous generator exists for time series. In contrast, a simple off-the-shelf classifier used as a black-box detector performs remarkably well, achieving an average F1 of 79.2, a 22.1% relative improvement over the white-box approach, and a TPR@1%FPR of 57.2. Diffusion-generated time series detection is therefore not a direct transfer of the image domain problem. This work provides the first systematic exploration of white-box and black-box detection for diffusion-generated time series. We close by identifying several open and promising directions.