ID-Crafter：面向组合式多主体视频生成的视觉语言模型接地在线强化学习
ID-Crafter: VLM-Grounded Online RL for Compositional Multi-Subject Video Generation
分层身份保持注意力（intra-subject/inter-subject/cross-modal 三级聚合）与 VLM 语义引导、在线 RL 微调的组合，可直接迁移到个性化视频生成与多主体可控生成中，解决多主体身份混淆与语义冲突问题。
原文依据 a hierarchical identity-preserving attention mechanism that progressively aggregates features at intra-subject, inter-subject, and cross-modal levels
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论高保真视频合成已取得显著进展，但现有范式往往难以有效整合来自多个主体的身份信息。这导致语义冲突，并在保持身份与交互方面表现欠佳，限制了可控性与适用性。为解决该问题，我们提出 ID-Crafter，一个面向多主体视频生成的框架，可实现更优的身份保持与语义一致性。ID-Crafter 集成了三个关键组件：(i) 分层身份保持注意力机制，在主体内、主体间与跨模态三个层级上逐步聚合特征；(ii) 由预训练视觉语言模型（VLM）驱动的语义理解模块，用于提供细粒度引导并捕捉复杂的主体间关系；(iii) 在线强化学习阶段，用于针对关键概念进一步精炼模型。此外，我们构建了一个新数据集，以支持稳健的训练与评测。大量实验表明，ID-Crafter 在多主体视频生成基准上取得了新的最优性能，在身份保持、时序一致性与整体视频质量方面均表现优异。
英文原文摘要 1275 字符
Significant progress has been achieved in high-fidelity video synthesis, yet current paradigms often fall short in effectively integrating identity information from multiple subjects. This leads to semantic conflicts and suboptimal performance in preserving identities and interactions, limiting controllability and applicability. To tackle this issue, we introduce ID-Crafter, a framework for multi-subject video generation that achieves superior identity preservation and semantic coherence. ID-Crafter integrates three key components: (i) a hierarchical identity-preserving attention mechanism that progressively aggregates features at intra-subject, inter-subject, and cross-modal levels; (ii) a semantic understanding module powered by a pretrained Vision-Language Model (VLM) to provide fine-grained guidance and capture complex inter-subject relationships; and (iii) an online reinforcement learning phase to further refine the model for critical concepts. Furthermore, we construct a new dataset to facilitate robust training and evaluation. Extensive experiments demonstrate that ID-Crafter establishes new state-of-the-art performance on multi-subject video generation benchmarks, excelling in identity preservation, temporal consistency, and overall video quality.