Spectral-Guided Diffusion: Accelerating Inference via Static Spectral Layer Scheduling

Diffusion inference repeatedly evaluates the same large network. We ask whether pretrained weights alone can identify residual branches that need not be recomputed throughout the trajectory. Our Spectral Concentration Ratio (SCR) measures leading-versus-tail singular-value energy. Combined with Frobenius magnitude, it yields an offline sensitivity proxy and a deterministic lifetime for each scheduled unit. A frozen unit reuses its cached residual-branch update while the current residual stream and all external conditioning continue to propagate. The method needs no router, calibration prompts, or input-dependent search. At matched layer-step budgets, SCR/Frobenius preserves quality better than random, depth, norm, stable-rank, and Frobenius--stable-rank schedules on LLaDA-8B, DiT-XL/2, U-ViT-L, and SDXL. Broader LLaDA tests cover retrieval, reasoning, code, summarization, and open-ended generation; matched-horizon controls retain the ranking down to ten denoising steps. The complete captured-graph system reaches 2.8×--3.0× wall-clock speedup over eager inference. This is a systems-level number: on LLaDA, padded graph execution already gives 2.7×, while eliminating inactive branch work raises it to 3.0×. The perturbation analysis motivates pre-norm attention and MLP components under explicit local assumptions; results on AdaLN, U-shaped, convolutional, and cross-attention blocks are empirical transfer, not certified guarantees.