One-Minute Video Generation with Test-Time Training

On a sunny morning in New York, Tom, a blue-gray cat carrying a briefcase, arrives at his office in the World Trade Center. As he settles in, his computer suddenly shuts down -Jerry, a mischievous brown mouse, has chewed the cable. A chase ensues, ending with Tom crashing into the wall as Jerry escapes into his mousehole. Determined, Tom bursts through an office door, accidentally interrupting a meeting led by Spike, an irritated bulldog, who angrily sends him away. Safe in his cozy mousehole, Jerry laughs at the chaos. Jerry happily eats cheese in a tidy kitchen until Tom playfully takes it away, teasing him. Annoyed, Jerry packs his belongings and leaves home, dragging a small suitcase behind him. Later, Tom notices Jerry's absence, feels sad, and follows Jerry's tiny footprints all the way to San Francisco. Jerry sits disheartened in an alleyway, where Tom finds him, gently offering cheese as an apology. Jerry forgives Tom, accepts the cheese, and the two return home together, their friendship restored. Figure 1. TTT layers enable pre-trained Diffusion Transformers to generate one-minute videos from text storyboards. We use Tom and Jerry cartoons as a proof-of-concept. The videos tell complete stories with coherent scenes composed of dynamic motion. Every video is produced directly by the model in a single shot, without editing, stitching, or post-processing. All stories are original.