Video DiT
PulseAugur coverage of Video DiT — every cluster mentioning Video DiT across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
New research enhances autonomous driving with realistic world models and data efficiency · 3 sources tracked
Researchers have developed new methods for improving autonomous driving systems by focusing on realistic world modeling and data efficiency. One approach, "Making the World React Realistically," couples behavior modelin…
-
LinMorph method learns weight transformation for efficient Video DiT linearization
Researchers have developed LinMorph, a novel method for efficiently linearizing pre-trained Video Diffusion Transformers (Video DiTs). Unlike previous approaches that focus on recovering quality after attention replacem…
-
New ForeWAM model predicts robot actions without future video decoding
Researchers have developed ForeWAM, a novel World Action Model (WAM) that enhances robot action generation by conditioning on predicted future states without requiring explicit future video decoding. This approach utili…
-
New EmoWorld framework offers controllable emotional video generation
Researchers have developed EmoWorld, a novel framework designed to enhance emotional control in video generation. This system decouples global atmosphere, semantic affect cues, and temporal progression, which are typica…