通过区分注意力头角色，实现分钟级视频生成
Head Forcing: Long Autoregressive Video Generation via Head Heterogeneity
- 按注意力头功能定制缓存策略，提升长视频生成效率
- 无需训练即可将生成时长从5秒扩展至分钟级
- 适合需要长序列生成与多提示交互的视频应用
自回归视频扩散模型支持实时合成，但在长序列中存在误差累积和上下文丢失问题。我们发现，自回归视频扩散变换器中的注意力头具有不同功能：局部头用于细节优化，锚点头用于结构稳定，记忆头用于远距离上下文聚合，但现有方法对此统一处理，导致键值缓存分配不佳。为此，我们提出无需训练的Head Forcing框架，为不同头类型分配定制化缓存策略：局部头和锚点头仅保留关键帧，记忆头则采用分层记忆系统并动态更新。同时引入头级旋转位置编码重编码机制，确保位置信息在预训练范围内。无需额外训练，该方法可将生成时长从5秒扩展至分钟级，支持多提示交互合成，并持续优于现有基线。
原文摘要 · Abstract (English)
Autoregressive video diffusion models support real-time synthesis but suffer from error accumulation and context loss over long horizons. We discover that attention heads in AR video diffusion transformers serve functionally distinct roles as local heads for detail refinement, anchor heads for structural stabilization, and memory heads for long-range context aggregation, yet existing methods treat them uniformly, leading to suboptimal KV cache allocation. We propose Head Forcing, a training-free framework that assigns each head type a tailored KV cache strategy: local and anchor heads retain only essential tokens, while memory heads employ a hierarchical memory system with dynamic episodic updates for long-range consistency. A head-wise RoPE re-encoding scheme further ensures positional encodings remain within the pretrained range. Without additional training, Head Forcing extends generation from 5 seconds to minute-level duration, supports multi-prompt interactive synthesis, and consistently outperforms existing baselines. Project Page: https://jiahaotian-sjtu.github.io/headforcing.github.io/.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。