AI 中文总结
Carnator通过从扩散模型内部状态提取生成原生兼容性证据，实现跨请求的缓存复用，在三个文本到视频骨干上达到2.17倍加速并保持生成质量。
AI 中文摘要
视频扩散Transformer能够生成高质量视频，但迭代去噪过程会产生大量推理延迟，限制了交互式和大规模服务。大多数现有的加速方法专注于单个请求，从而将效率提升限制在单条生成轨迹内部的冗余上。最近的跨请求复用提供了额外的节省来源，但现有方法通常从粗粒度的语义相似性推断可复用性。这混淆了语义相关性与生成级计算兼容性，因此激进的复用可能接受不兼容的历史计算，而保守的复用则留下大量未实现的加速。我们提出Carnator，一个跨请求加速框架，通过直接从模型演化的内部状态中提取和使用生成原生兼容性证据来解决这一挑战。具体而言，Carnator执行一次轻量级早期探测，从内部扩散状态构建早期签名，通过风险感知的兼容性决策评估复用有效性。相同的证据通过定位目标特定计算来表征复用范围，并指导历史潜轨迹和稀疏注意力连接性的联合复用。在三个文本到视频骨干网络上，尽管缓存接受更具选择性，Carnator始终比评估的跨请求基线实现更高的缓存命中端到端加速，达到高达2.17倍加速，同时保持有竞争力的生成质量。
英文摘要
Video diffusion transformers produce high-quality videos, yet iterative denoising incurs substantial inference latency, limiting interactive and large-scale serving. Most existing acceleration methods focus on individual requests, thereby restricting efficiency gains to redundancy within a single generation trajectory. Recent cross-request reuse offers an additional source of savings, but existing approaches often infer reusability from coarse semantic similarity. This conflates semantic relatedness with generation-level computational compatibility, so aggressive reuse may accept incompatible historical computation while conservative reuse leaves substantial acceleration unrealized. We present \emph{Carnator}, a cross-request acceleration framework that addresses this challenge by extracting and using generation-native compatibility evidence directly from the model's evolving internal states. Specifically, \emph{Carnator} performs a lightweight early probe to construct an Early Signature from internal diffusion states, assessing reuse validity through risk-aware compatibility decisions. The same evidence characterizes reuse scope by localizing target-specific computation and guiding joint reuse of historical latent trajectories and sparse attention connectivity. Across three text-to-video backbones, Carnator consistently achieves higher cache-hit end-to-end acceleration than the evaluated cross-request baselines despite more selective cache acceptance, reaching up to 2.17$\times$ speedup while maintaining competitive generation quality.
