Tag
MV-Forcing introduces a diffusion framework that combines temporal and view-wise autoregression to generate long, multi-view consistent videos of dynamic scenes, using a 4D geometric bridge and spatio-temporal distillation to enable arbitrary-length generation from a few-step student model.
This article explains why AI tokens are expensive: the autoregressive generation process requires predicting token by token and repeatedly computing the entire context, causing computational cost to grow linearly with context length; coding agents, due to multi-turn interactions and file reads, quickly accumulate very long contexts, further increasing token costs.
Memento is a subject-reconstruction-guided framework that improves long-form video generation by preserving recurring subjects through memory-based reconstruction and dual-query mechanisms, achieving state-of-the-art performance in long-term subject consistency and cross-shot coherence.
TBD-VLA introduces a discrete vision-language-action framework that combines block diffusion with autoregressive generation to achieve efficient temporal action modeling and faster inference, significantly outperforming prior VLA approaches in simulation and real-world manipulation tasks.
This academic paper develops a theoretical framework for online learning with autoregressive chain-of-thought reasoning, analyzing mistake bounds under end-to-end and trajectory supervision models.