Spectral Tail Auxiliary Learning for AI-Generated Image Detection
About
As generative image models evolve rapidly, the perceptual gap between generated and real images continues to narrow, making AI-generated image detection increasingly challenging. Many existing methods exploit frequency-domain cues for detection, typically described as frequency-domain artifacts or high-frequency discrepancies. However, the specific and recurring spectral regularities remain insufficiently understood and characterized. In this paper, we systematically analyze the one-dimensional radial log-power spectra of real and generated images. We find that generated images do not necessarily exhibit higher or lower energy across the entire spectrum or high-band range. Instead, their spectra deviate from the power-law decay and show an anomalous uplift in the ultra-high-frequency tail. We term this phenomenon spectral tail uplift. We further attribute this phenomenon to nonlinear harmonic accumulation in trained generative models, suggesting that it can serve as a structural cue across generative architectures. Based on this observation, we propose Spectral Tail Auxiliary Learning (STAL), a frequency-domain auxiliary supervision framework for generalizable AI-generated image detection. STAL transfers spectral-tail cues from a tail-aware frequency teacher to a spatial detector during training, while all frequency-domain modules are discarded at inference time. Consequently, STAL introduces no inference overhead. Extensive experiments on 9 public datasets show that STAL achieves strong generalization and stability across generators, data distributions, and real-world scenarios.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| AI-generated image detection | GenImage | Midjourney Detection Rate98.2 | 154 | |
| Synthetic Image Detection | ForenSynths (test) | Mean Accuracy92.6 | 60 | |
| Synthetic Image Detection | DRCT-2M | Average Score99.8 | 57 | |
| AIGC Detection | AIGCDetectBenchmark | -- | 50 | |
| AIGI Detection | AIGCDetect | B.Acc96 | 46 | |
| AIGI Detection | SynthWildx | DALLE3 Performance Score95.6 | 46 | |
| AIGI Detection | Synthbuster | B.Acc99.7 | 35 | |
| AIGI Detection | WildRF | FB Score98.4 | 23 | |
| AI-generated image detection | Synthbuster | DALL·E 2 Score99.6 | 23 | |
| AIGC Detection | SynthWildx | Balanced Accuracy95.2 | 23 |