RealtimeWAM: One-Step Asynchronous World Action Models

arXiv:2610.06617v1 Announce Type: new Abstract: World Action Models (WAMs) incorporate visual representations from video generation backbones to guide action prediction. Recent efficient WAMs adopt Mixture-of-Transformers (MoT) architectures and compute video representations once for reuse by the action expert. However, intra-expert iteration (ie, multi-step action denoising) and inter-expert waiting (ie, sequential execution of the video and action experts) still limit inference efficiency. To this end, we present RealtimeWAM, an extremely efficient WAM variant with one-step action generation and asynchronous inference, addressing these two bottlenecks. To reduce intra-expert iteration, we propose Teacher-Anchored Consistency Distillation (TACD) to address a local-global error gap: low local consistency error alone does not guarantee accurate final actions. TACD supplements local consistency with explicit supervision from the frozen teacher's multi-step rollout endpoint, enabling accurate one-step action generation. Additionally, we propose Cross-Expert Wavefront Pipelining (CEWP) to eliminate unnecessary expert-level waiting. It overlaps the two experts through block-wise sharing of the video KV cache, synchronizing only immediately before the corresponding action attention consumes it. Extensive experiments across diverse benchmarks (eg, LIBERO, LIBERO-Plus and RoboTwin) and model variants (eg, Fast-WAM and Faster-WAM) demonstrate the superiority of RealtimeWAM. Notably, RealtimeWAM maintains near-lossless performance (ie, <1% drop) across these benchmarks while delivering significant end-to-end speedup (eg, sim25imes on H100). Our code and checkpoints are available via this href{https://github.com/ModelTC/LightX2V/tree/main/examples/realtimewam}{link}.

Source: arXiv cs.CV | 2026-10-06

Related