Megadose AI progress, ranked and analyzed.

Structured Residual Connectivity Matters for Diffusion Transformers

· HF Daily Papers ·
The paper argues DiTs train and synthesize better when residual paths actively retrieve earlier features instead of treating depth as one pooled stream.

The authors report that DiT representations show a preference for early-layer reuse and symmetric layer guidance. Their proposed connectivity adds local residual links plus long-range cross-depth paths, letting blocks attend to earlier spatial and semantic cues. In experiments, the approach cut training iterations by up to 1.73x and added under 0.1% parameters. It improved a REPA-XL/2 model from 5.9 to 4.34 FID without guidance, and reached 1.39 FID with classifier-free guidance. HF Daily Papers' note

score 4

Categories: Research