2024/10/30 by Yining Hong, Hong, Yining, Baiyu Liu +21 · 12 citations
Biochemistry, Genetics and Molecular Biology · Computer Science · #Advanced Vision and Imaging #Artificial Intelligence (cs.AI) #Cell Image Analysis Techniques #Computation and Language (cs.CL) #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Human Pose and Action Recognition #Machine Learning (cs.LG) #Robotics (cs.RO)
paper · pdf · doi:10.48550/arxiv.2410.23277
openalex publication_date 2024/10/30 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Human beings are endowed with a complementary learning system, which bridges the slow learning of general world dynamics with fast storage of episodic memory from a new experience. Previous video generation models, however, primarily focus on slow learning by pre-training on vast amounts of data, overlooking the fast learning phase crucial for episodic memory storage. This oversight leads to inconsistencies across temporally distant frames when generating longer videos, as these frames fall beyond the model's context window. To this end, we introduce SlowFast-VGen, a novel dual-speed learning system for action-driven long video generation. Our approach incorporates a masked conditional video diffusion model for the slow learning of world dynamics, alongside an inference-time fast learning strategy based on a temporal LoRA module. Specifically, the fast learning process updates its temporal LoRA parameters based on local inputs and outputs, thereby efficiently storing episodic memory in its parameters. We further propose a slow-fast learning loop algorithm that seamlessly integrates the inner fast learning loop into the outer slow learning loop, enabling the recall of prior multi-episode experiences for context-aware skill learning. To facilitate the slow learning of an approximate world model, we collect a large-scale dataset of 200k videos with language action annotations, covering a wide range of scenarios. Extensive experiments show that SlowFast-VGen outperforms baselines across various metrics for action-driven video generation, achieving an FVD score of 514 compared to 782, and maintaining consistency in longer videos, with an average of 0.37 scene cuts versus 0.89. The slow-fast learning loop algorithm significantly enhances performances on long-horizon planning tasks as well. Project Website: https://slowfast-vgen.github.io