SIGMA：基于多属性 token 的选择性交错生成
SIGMA: Selective-Interleaved Generation with Multi-Attribute Tokens
SIGMA 提出在扩散 Transformer 中通过选择性多属性 token 实现交错多条件生成，为视觉研究者提供了一种可复用的后训练框架，能灵活组合风格、内容、主体和身份等多种异构条件，提升编辑任务的可控性与跨条件一致性。
原文依据 SIGMA introduces selective multi-attribute tokens, including style, content, subject and identity tokens, which allow the model to interpret and compose multiple visual conditions in an interleaved text-image sequence.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论近期 Bagel 等统一模型表明，成对的图像编辑数据能够在一个扩散 Transformer 内有效对齐多种视觉任务。然而，这些模型仍局限于单条件输入，缺乏从多种异构来源合成结果的灵活性。我们提出 SIGMA（基于多属性 token 的选择性交错生成），这是一个统一的后训练框架，能够在扩散 Transformer 内实现交错的多条件生成。SIGMA 引入选择性多属性 token，包括风格、内容、主体和身份 token，使模型能够在交错的文本-图像序列中解释并组合多种视觉条件。通过在 Bagel 统一骨干网络上使用 70 万交错样本进行后训练，SIGMA 支持组合式编辑、选择性属性迁移和细粒度多模态对齐。大量实验表明，SIGMA 在多种编辑和生成任务中提升了可控性、跨条件一致性和视觉质量，在组合式任务上相比 Bagel 取得显著增益。代码见 https://github.com/auihund/SIGMA。
英文原文摘要 1189 字符
Recent unified models such as Bagel demonstrate that paired image-edit data can effectively align multiple visual tasks within a single diffusion transformer. However, these models remain limited to single-condition inputs and lack the flexibility needed to synthesize results from multiple heterogeneous sources. We present SIGMA (Selective-Interleaved Generation with Multi-Attribute Tokens), a unified post-training framework that enables interleaved multi-condition generation within diffusion transformers. SIGMA introduces selective multi-attribute tokens, including style, content, subject and identity tokens, which allow the model to interpret and compose multiple visual conditions in an interleaved text-image sequence. Through post-training on the Bagel unified backbone with 700K interleaved examples, SIGMA supports compositional editing, selective attribute transfer and fine-grained multimodal alignment. Extensive experiments show that SIGMA improves controllability, cross-condition consistency and visual quality across diverse editing and generation tasks, with substantial gains over Bagel on compositional tasks. Code is available at https://github.com/auihund/SIGMA.