GeoShrink: Accelerating Diffusion Transformers with Two Lines of Code

arXiv:2609.33723v1 Announce Type: new Abstract: Diffusion transformers incur substantial inference cost through repeated model evaluations along a sampling trajectory. We introduce GeoShrink, a training-free acceleration method that retains the original solver grid while evaluating the model only at a prescribed set of anchors. At skipped stages, GeoShrink predicts the solver-facing output by adding a geometrically retained fraction of the latest observed innovation to the most recent exact output. We derive this rule from chordal tangent transport and round-trip line projection, and establish a geometric anchor-spacing principle that minimizes the largest adjacent gap expansion under fixed coverage and first span. The analysis characterizes the geometric closure and propagation of prediction errors without assuming access to future model outputs. Experiments cover image, video, motion, and audio generation, together with adapted 3D backends. At approximately 5imes acceleration, GeoShrink improves FLUX PSNR by 3.10 dB over the strongest listed baseline. On HunyuanVideo, it achieves a reported 4.99imes speedup and improves ChronoMagic-Bench-150 PSNR by 5.44 dB over the strongest listed fidelity baseline. Comparisons at fixed evaluation budgets further show substantial gains on motion, audio, music, and 3D generation.

Source: arXiv cs.CV | 2026-09-29