SOURCE-LINKED INTELLIGENCE

DART: Distillation-Aware Reparameterization for Training-Free LoRA Reuse in Few-Step Video Diffusion Models

arXiv · Artificial Intelligence · article · Sep 17, 2026 · UTC

Step distillation reduces the cost of video generation, but reusing a LoRA trained for a longer trajectory can alter its functional effect or degrade target quality. Static parameter compatibility offers one perspective on this problem; our observations show that similar measured geometry can coexist with different adapter behavior under a shortened denoising schedule. We propose DART, a training-free method that combines low-rank coordinate transport with target-schedule response calibration using forward evaluations and no source training videos. On a four-step Wan2.2 target, DART-F improves

Read original source ↗ Open in workspace

recordType

paper

region

Global