Autoregressive video generation has improved rapidly in visual fidelity and interactivity, but it still suffers from long-term inconsistency and memory degradation. Most existing solutions either compress historical frames using predefined strategies or retrieve keyframes based on coarse implicit attention signals, both of which fail to handle evolving prompts with shifting entity references, leading to identity drift, character duplication, and attribute loss. To address this, we propose IAMFlow, a training-free identity-aware memory framework that explicitly models and tracks persistent entity identities, enabling consistent generation across prompt transitions. Specifically, an LLM extracts entities with visual attributes from each prompt and assigns unique global IDs for identity-aware memory, while a VLM asynchronously verifies and refines attributes from rendered frames, enabling explicit entity tracking in place of implicit similarity-based matching. To keep the proposed framework computationally practical, we design a systematic inference acceleration pipeline, including asynchronous visual verification, adaptive prompt transition, and model quantization, which achieves faster generation than existing baselines. Furthermore, we introduce NarraStream-Bench, a benchmark for narrative streaming video generation that features 324 multi-prompt scripts spanning six dimensions and a three-dimensional evaluation protocol that integrates both traditional metrics and multimodal large language model-based assessments. Extensive experiments show that IAMFlow, despite being training-free, achieves the best overall performance on NarraStream-Bench, outperforming the strongest baseline by 2.56 points, while achieving a 1.39$\times$ speedup over the most efficient baseline in the 60-second multi-prompt setting.

翻译：自回归视频生成在视觉逼真度和交互性方面取得了快速进展，但仍存在长期不一致性和记忆退化问题。现有方法大多采用预定义策略压缩历史帧，或基于粗粒度的隐式注意力信号检索关键帧——两者均无法有效应对包含动态实体引用的演进式提示，导致身份漂移、角色重复和属性丢失。为此，我们提出IAMFlow——一种无需训练的身份感知记忆框架，通过显式建模和追踪持久化实体身份，实现提示过渡下的一致性生成。具体而言，大语言模型（LLM）从每个提示中提取具有视觉属性的实体，并为身份感知记忆分配唯一全局标识符；视觉语言模型（VLM）则异步验证并优化渲染帧中的属性，以显式实体追踪替代隐式相似度匹配。为保持框架的计算实用性，我们设计了系统化的推理加速流程，包括异步视觉验证、自适应提示过渡和模型量化，在生成速度上超越现有基线。此外，我们提出NarraStream-Bench——面向叙事流式视频生成的基准测试，包含覆盖六个维度的324组多提示脚本，以及融合传统指标与多模态大语言模型评估的三维评价协议。大量实验表明，IAMFlow尽管无需训练，仍在NarraStream-Bench上取得综合最优性能，比最强基线高出2.56分，并在60秒多提示设定下实现相比最高效基线1.39倍的加速。