MM-DiT
PulseAugur coverage of MM-DiT — every cluster mentioning MM-DiT across labs, papers, and developer communities, ranked by signal.
3 day(s) with sentiment data
-
PC-Edit framework enhances image editing with prompt-contrastive region discovery
Researchers have introduced PC-Edit, a novel prompt-contrastive framework designed for training-free editing in MM-DiT models. This method directly contrasts image-token attention outputs under source and target prompts…
-
Humanoid robot model Ψ₀ trains effectively on limited real-world data
Researchers at the University of Southern California's PSI Lab have developed a new foundation model for humanoid robots called Ψ₀, which focuses on optimizing the use of limited real-world robot data. Instead of relyin…
-
New method steers video generation models without retraining
Researchers have developed a method to control video generation models without retraining the core model. By injecting differentiable energy guidance during the sampling phase, they can steer the planned trajectory of a…
-
New method enables text-and-image-to-image generation without retraining
Researchers have developed TF-TI2I, a novel method for text-and-image-to-image generation that adapts existing text-to-image models without requiring further training. This approach leverages the MM-DiT architecture, en…
-
New method enables open-vocabulary scene text editing with style consistency
Researchers have developed a novel self-prompting method for editing scene text in images, addressing limitations of existing approaches that neglect visual details of target regions and are constrained by pre-trained g…
-
New method improves AI portrait generation by balancing alignment, realism, and aesthetics
Researchers have developed a new method to improve human portrait generation in text-to-image diffusion models, addressing the common trade-offs between text-image alignment, realism, and aesthetics. Their approach uses…
-
Galaxy General LDA-1B model unifies diverse data for embodied AI's GPT-2 moment
Galaxy General LDA has introduced LDA-1B, a 1.6 billion parameter model designed to unify the utilization of diverse data sources for embodied AI. This model employs a novel World-Action Fusion approach, enabling it to …
-
UniSonate model unifies speech, music, and sound effect generation
Researchers have developed UniSonate, a novel unified framework for generating speech, music, and sound effects using natural language instructions. This model addresses the fragmentation in generative audio by reconcil…