Search papers, labs, and topics across Lattice.
ShotStream introduces a causal architecture for multi-shot video generation, enabling interactive storytelling with low latency. They fine-tune a text-to-video model into a next-shot generator and distill it into a causal student using Distribution Matching Distillation, incorporating a dual-cache memory mechanism and RoPE discontinuity indicator for visual coherence. Results show ShotStream achieves 16 FPS on a single GPU while matching or exceeding the quality of slower bidirectional models.
Generate multi-shot videos at 16 FPS with a single GPU and interactively steer the narrative in real-time, thanks to a novel causal architecture that overcomes the limitations of bidirectional models.
Multi-shot video generation is crucial for long narrative storytelling, yet current bidirectional architectures suffer from limited interactivity and high latency. We propose ShotStream, a novel causal multi-shot architecture that enables interactive storytelling and efficient on-the-fly frame generation. By reformulating the task as next-shot generation conditioned on historical context, ShotStream allows users to dynamically instruct ongoing narratives via streaming prompts. We achieve this by first fine-tuning a text-to-video model into a bidirectional next-shot generator, which is then distilled into a causal student via Distribution Matching Distillation. To overcome the challenges of inter-shot consistency and error accumulation inherent in autoregressive generation, we introduce two key innovations. First, a dual-cache memory mechanism preserves visual coherence: a global context cache retains conditional frames for inter-shot consistency, while a local context cache holds generated frames within the current shot for intra-shot consistency. And a RoPE discontinuity indicator is employed to explicitly distinguish the two caches to eliminate ambiguity. Second, to mitigate error accumulation, we propose a two-stage distillation strategy. This begins with intra-shot self-forcing conditioned on ground-truth historical shots and progressively extends to inter-shot self-forcing using self-generated histories, effectively bridging the train-test gap. Extensive experiments demonstrate that ShotStream generates coherent multi-shot videos with sub-second latency, achieving 16 FPS on a single GPU. It matches or exceeds the quality of slower bidirectional models, paving the way for real-time interactive storytelling. Training and inference code, as well as the models, are available on our
