Search papers, labs, and topics across Lattice.
This paper introduces Ring Forcing, an autoregressive video diffusion framework that addresses the limitations of long-term memory in video generation by enhancing both object permanence and memory capacity. The proposed ring-structured training strategy facilitates effective retrieval from distant historical context, balancing the need for strict adherence to past information with the diversity of generated content. Experimental results show that Ring Forcing significantly improves coherence and object permanence over extended durations, outperforming existing state-of-the-art methods in video generation.
Achieving minutes-long coherence in video generation, Ring Forcing reconciles the trade-off between historical fidelity and generative diversity.
Scaling video generation to long durations reveals a critical bottleneck: current models lack robust long-term memory. This deficiency can be studied along two critical aspects: object permanence, the ability to precisely reproduce the appearance of objects upon re-entry; and memory capacity, the ability to process ultra-long context and use information from distant history. Robust long-term memory requires both: object permanence without sufficient context handling limits the temporal scope, while long context length without permanence fails to maintain identity. To address this, we present Ring Forcing, an autoregressive video diffusion framework designed to robustly construct and precisely utilize long-term memory. Our ring-structured training strategy enforces retrieval from distant history, effectively reconciling the trade-off between strict historical adherence and generative diversity. To expand memory capacity, we introduce a compression and timestep composition strategy. Under fixed sequence length constraints, this method extends the effective historical span to minutes-long durations and achieves a comprehensive receptive field over the entire history. Furthermore, we present a sparse RoPE mechanism to enable flexible, scalable memory adaptation while fully exploiting pre-trained priors. Extensive experiments demonstrate that Ring Forcing achieves superior minutes-long coherence and object permanence, significantly outperforming state-of-the-art methods.
