SD3.5M
PulseAugur coverage of SD3.5M — every cluster mentioning SD3.5M across labs, papers, and developer communities, ranked by signal.
-
New technique suppresses image-to-text feedback to improve AI image generation
Researchers have developed a new technique called Semantic Feedback Guidance (SFG) to improve the quality of images generated by multimodal diffusion transformers (MMDiTs). They found that the feedback loop from image-t…
-
New Truncated Jump Sampling accelerates AI image generation without retraining
Researchers have introduced a novel method called Truncated Jump Sampling (TJS) to accelerate the generation process in diffusion and flow matching models. This technique, based on the concept of 'endpoint decodability'…
-
TrioPose framework enhances multi-person image generation with diffusion transformers
Researchers have developed TrioPose, a novel framework for pose-guided text-to-image generation that addresses challenges in complex multi-person scenarios. Built on the SD3.5M architecture, TrioPose utilizes a Triple-S…