AI-SOVEREIGN
Home / AI Research / VAE Rhythm: Temporal Decoding Traces for AI-Generated Video Detection
VAAI Research

VAE Rhythm: Temporal Decoding Traces for AI-Generated Video Detection

by showlab · Ke Sun, Shuwei Li, Haiyang Mei, Mike Zheng Shou

Visible artifacts are an unreliable cue for detecting generated videos, because newer generators learn to remove them. We instead detect a trace left by the generation pipeline itself. Most video generators decode latents with a 3D causal VAE that groups consecutive frames in time, so the fidelity of generated frames rises and falls periodically with their position in a group. We call this periodic trace VAE Rhythm. Real videos are captured frame by frame and do not have it. To reveal it without training, we propose a phase scan. The scan reconstructs a clip with an off-the-shelf VAE from successive start offsets and measures the periodicity of the reconstruction error across these…

Upvotes6
Comments2
Published15 Nov 2026

More from showlab