何时锁定注意力：视频扩散中的无训练KV控制
When to Lock Attention: Training-Free KV Control in Video Diffusion
AI总结：
KV-Lock通过动态调度KV融合比和CFG尺度，提升视频扩散模型中前景质量与背景一致性的平衡，实现无训练的高效视频编辑。
摘要：
在视频编辑中，保持背景一致性的同时提升前景质量仍然是核心挑战。注入完整图像信息往往会导致背景伪影，而刚性背景锁定严重限制了模型对前景生成的能力。为了解决这一问题，我们提出KV-Lock，一种专为基于DiT的视频扩散模型设计的无训练框架。我们的核心见解是，幻觉度量（去噪预测的方差）直接量化生成多样性，这与无分类器指导（CFG）尺度内在相关。在此基础上，KV-Lock利用扩散幻觉检测动态调度两个关键组件：缓存背景键值（KVs）与新生成KVs之间的融合比，以及CFG尺度。当检测到幻觉风险时，KV-Lock增强背景KV锁定，并同时放大对前景生成的条件指导，从而减少伪影并提高生成保真度。作为无训练、即插即用的模块，KV-Lock可以轻松集成到任何预训练的DiT模型中。广泛实验验证了我们的方法在多种视频编辑任务中，在提高前景质量的同时保持高背景保真度方面优于现有方法。
英文摘要：
Maintaining background consistency while enhancing foreground quality remains a core challenge in video editing. Injecting full-image information often leads to background artifacts, whereas rigid background locking severely constrains the model's capacity for foreground generation. To address this issue, we propose KV-Lock, a training-free framework tailored for DiT-based video diffusion models. Our core insight is that the hallucination metric (variance of denoising prediction) directly quantifies generation diversity, which is inherently linked to the classifier-free guidance (CFG) scale. Building upon this, KV-Lock leverages diffusion hallucination detection to dynamically schedule two key components: the fusion ratio between cached background key-values (KVs) and newly generated KVs, and the CFG scale. When hallucination risk is detected, KV-Lock strengthens background KV locking and simultaneously amplifies conditional guidance for foreground generation, thereby mitigating artifacts and improving generation fidelity. As a training-free, plug-and-play module, KV-Lock can be easily integrated into any pre-trained DiT-based models. Extensive experiments validate that our method outperforms existing approaches in improved foreground quality with high background fidelity across various video editing tasks.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
