Abstract
World-action models (WAMs) couple predictive visual modeling with action generation, typically relying on iterative denoising with a fixed denoising steps. However, manipulation tasks contain actions chunks with varying sensitivity to generation errors: critical actions require precision, while less sensitive actions allow faster generation with fewer denoising steps. Here we introduce AnyStep World Action Model, a general framework for tunable-budget prediction and scene-dependent computation allocation. Our budget-aligned teacher-trajectory distillation trains interval-conditioned flow maps using explicit frozen-teacher transitions and shared low-rank adapters, supporting action generation from one-step prediction to multi-step refinement. Building on this capability, a lightweight risk-benefit scheduler predicts teacher-curvature-based difficulty and budget-specific student-teacher fidelity from a single one-step preview, selecting the smallest budget predicted to satisfy risk-adaptive fidelity requirements. We evaluate our framework on three widely used WAMs Motus, FastWAM, and LingBotVA using RoboTwin 2.0. Our method reduces average denoising steps by 60.2%, 49.8%, and 85.28%, respectively, while maintaining baseline task success rates. In particular, our AnyStep training substantially improves model performance under a one-step denoising budget, increasing task success rates by 7.07%, 12.08%, and 8.94% on Motus, FastWAM, and LingBotVA, respectively. Experiments on six real-world manipulation tasks further validate its effectiveness.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Wang, R., Wang, X., Yang, D., Li, Y., Chen, C., Wang, Z., & Qi, X. (2026). AnyStep-WAM: Budget-Aligned Distillation and Adaptive Inference for World Action Models. https://omanscience.com/en/articles/anystep-wam-budget-aligned-distillation-and-adaptive-inference-for-world-action-models
MLA 9
Wang, Rui, et al. "AnyStep-WAM: Budget-Aligned Distillation and Adaptive Inference for World Action Models." https://omanscience.com/en/articles/anystep-wam-budget-aligned-distillation-and-adaptive-inference-for-world-action-models.
Chicago (author–date)
Wang, Rui, Xiangyu Wang, Donglin Yang, Yibo Li, Canyang Chen, Zhongrui Wang, and Xiaojuan Qi. 2026. "AnyStep-WAM: Budget-Aligned Distillation and Adaptive Inference for World Action Models." https://omanscience.com/en/articles/anystep-wam-budget-aligned-distillation-and-adaptive-inference-for-world-action-models.
Harvard
Wang, R., Wang, X., Yang, D., Li, Y., Chen, C., Wang, Z. and Qi, X. (2026) 'AnyStep-WAM: Budget-Aligned Distillation and Adaptive Inference for World Action Models', Available at: https://omanscience.com/en/articles/anystep-wam-budget-aligned-distillation-and-adaptive-inference-for-world-action-models.
Vancouver
Wang R, Wang X, Yang D, Li Y, Chen C, Wang Z, et al. AnyStep-WAM: Budget-Aligned Distillation and Adaptive Inference for World Action Models. https://omanscience.com/en/articles/anystep-wam-budget-aligned-distillation-and-adaptive-inference-for-world-action-models
IEEE
R. Wang, X. Wang, D. Yang, Y. Li, C. Chen, Z. Wang, and X. Qi, "AnyStep-WAM: Budget-Aligned Distillation and Adaptive Inference for World Action Models," https://omanscience.com/en/articles/anystep-wam-budget-aligned-distillation-and-adaptive-inference-for-world-action-models.