From SRA to Self-Flow: Data Augmentation or Self-Supervision?
Representation alignment has become an effective way to accelerate diffusion transformer training and improve generation quality. Recent self-alignment methods, such as SRA and Self-Flow, further remove the dependency on external pretrained encoders by constructing alignment within the diffusion model itself. However, the mechanism behind the improvement from SRA to Self-Flow, dual-time scheduling, remains under-examined: Self-Flow attributes its gain to interactions between tokens at different noise levels, where cleaner tokens help infer noisier ones. In this work, we revisit this explanatio
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
- Linked via arxiv authorDengyang Jiang →
From SRA to Self-Flow: Data Augmentation or Self-Supervision?
- Linked via arxiv authorMengmeng Wang →
From SRA to Self-Flow: Data Augmentation or Self-Supervision?
- Linked via arxiv authorHarry Yang →
From SRA to Self-Flow: Data Augmentation or Self-Supervision?
- Linked via arxiv authorJingdong Wang →
From SRA to Self-Flow: Data Augmentation or Self-Supervision?