SIGMA: Selective-Interleaved Generation with Multi-Attribute Tokens

TL;DR

SIGMA is a novel diffusion transformer framework that enables flexible, multi-attribute, and interleaved multi-condition image generation and editing, significantly enhancing controllability and consistency over previous models.

Contribution

It introduces selective multi-attribute tokens and a post-training framework for interleaved multi-condition image synthesis within diffusion transformers.

Findings

Improves controllability and cross-condition consistency.

Enhances visual quality across diverse tasks.

Achieves substantial gains over prior models like Bagel.

Abstract

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.