UniCP: A Unified Caching and Pruning Framework for Efficient Video Generation
This addresses computational bottlenecks in video generation for AI researchers and practitioners, though it appears incremental as it builds on existing caching methods.
The paper tackles the computational inefficiency of Diffusion Transformers in video generation by proposing UniCP, a unified caching and pruning framework that dynamically adjusts cache windows and prunes redundant attention components. Experimental results show UniCP outperforms existing methods in performance and efficiency.
Diffusion Transformers (DiT) excel in video generation but encounter significant computational challenges due to the quadratic complexity of attention. Notably, attention differences between adjacent diffusion steps follow a U-shaped pattern. Current methods leverage this property by caching attention blocks, however, they still struggle with sudden error spikes and large discrepancies. To address these issues, we propose UniCP a unified caching and pruning framework for efficient video generation. UniCP optimizes both temporal and spatial dimensions through. Error Aware Dynamic Cache Window (EDCW): Dynamically adjusts cache window sizes for different blocks at various timesteps, adapting to abrupt error changes. PCA based Slicing (PCAS) and Dynamic Weight Shift (DWS): PCAS prunes redundant attention components, and DWS integrates caching and pruning by enabling dynamic switching between pruned and cached outputs. By adjusting cache windows and pruning redundant components, UniCP enhances computational efficiency and maintains video detail fidelity. Experimental results show that UniCP outperforms existing methods in both performance and efficiency.