: LCT expands the context window of video diffusion models, allowing them to maintain visual and dynamic consistency across multiple different shots within a single scene.
: It facilitates a "director-like" workflow where users can progressively develop content shot-by-shot, using previously generated footage as a reference for the next segment. lu-VEnnv2a4.mp4
: By using context-causal attention, the model supports efficient long-form generation, making it practical for creating extended visual narratives. : LCT expands the context window of video