LiveAct: Towards Hour-Scale Real-Time Human Animation with Neighbor Forcing and ConvKV Memory
TL;DR
Neighbor Forcing is proposed, a diffusion-step-consistent AR formulation that propagates temporally adjacent frames as latent neighbors under the same noise condition that provides a distribution-aligned and stable learning signal throughout the AR chain, and a structured ConvKV Memory mechanism that compresses the keys and values in causal attention into a fixed-length representation, enabling constant-memory inference.
Abstract
Autoregressive (AR) diffusion models offer a promising framework for sequential generation such as video synthesis by combining diffusion modeling with causal inference. Although they support streaming generation, existing AR diffusion methods struggle to scale efficiently. In this paper, we identify two key challenges in hour-scale real-time human animation. First, most forcing strategies propagate sample-level representations with mismatched diffusion states, causing inconsistent learning signals and unstable convergence. Second, historical representations grow unbounded and lack structure, preventing effective reuse of cached states and severely limiting inference efficiency. To address these challenges, we propose Neighbor Forcing, a diffusion-step-consistent AR formulation that propagates temporally adjacent frames as latent neighbors under the same noise condition. This design provides a distribution-aligned and stable learning signal throughout the AR chain. Building upon this formulation, we introduce a structured ConvKV Memory mechanism that compresses the keys and values in causal attention into a fixed-length representation, enabling constant-memory inference and effectively unbounded video generation without short-term motion-frame memory. Experiments show improved convergence, hour-scale generation quality, and inference efficiency over existing AR diffusion methods. SoulX-LiveAct enables hour-scale real-time human animation at 20 FPS on two NVIDIA H100 or H200 GPUs while attaining strong lip-sync accuracy, human animation quality, emotional expressiveness, and inference efficiency.