This work systematically compares six causal structures, eight latent representations, and four training objectives, covering popular design choices in existing WAMs, to provide a systematic understanding of how core design choices affect world-action modeling and what principles can guide the development of future WAM systems.
Abstract
World Action Models (WAMs) have emerged as a promising paradigm for generalizable robot control. Despite the growing number of WAM systems, existing works often introduce unified systems that bundle together multiple design choices, such as architecture and training strategy, making it difficult to isolate individual contributions and systematically compare alternative designs. In this work, we present a controlled study that disentangles these design choices and analyzes not only their empirical effects, but also how and why they shape WAMs. More specifically, we focus on three fundamental questions in building WAMs: (1) what causal structure should govern the interaction between world modeling and action generation? (2) in which latent space should world modeling be performed? and (3) how do different world-action modeling objectives affect model behavior and performance? Through structurally controlled experiments on three representative benchmarks, RoboCasa-GR1, LIBERO, and LIBERO-Plus, we systematically compare six causal structures, eight latent representations, and four training objectives, covering popular design choices in existing WAMs. We further validate our key findings on real-robot data from the DROID dataset. We hope to provide a systematic understanding of how core design choices affect world-action modeling and what principles can guide the development of future WAM systems.
Align Once, Evaluate Everywhere (AoE), a camera-space inverse-dynamics model pretrained on large-scale robot data, is introduced, which provides this recovery interface for different future sources across tasks and environments and raises the execution success of generated futures from near zero to 47.0%.
Chen Zhao, Zi-Peng Ye, Yi-Han Lin et al.· 0 citations
This work studies action-conditioned predictors built on five frozen self-supervised visual backbones, finding that a world model guides action selection reliably only when the goal lies within, or slightly beyond, the trajectory it imagines during planning.
A. Alrasheed, Basim Azam, Naveed Akhtar· arXiv.org· 3 citations
World-action models (WAMs) couple future prediction with robot control, yet existing systems often vary the video backbone, interaction structure, supervision, and inference procedure simultaneously, making their design choices difficult to compare. We introduce OPENWAM, an open world-action modeling framework built ar...
Heng Yu, David Yuan, Ju-Ze Zhang et al.· 3 citations· ⚡1
World action models (WAMs) jointly generate actions and imagined futures, but a plausible prediction need not be realized by its co-generated action. We introduce WISE (\emph{World Action Inference Scaling through Evaluation}), a pre-execution selector that scores each sampled rollout on action validity and temporal re...
C. Ahmed, Mahdi Mostajabdaveh, M. Yazdani et al.· 0 citations
World and action modeling are closely related, yet their reciprocal benefits under unified causal training remain unclear. We introduce OneWM, a simple framework for unified causal training that treats actions as both conditions for world prediction and targets for policy learning within a shared video-action architect...
Yue Su, Bo-Kai Lin, Xin-Kai Wang et al.· 1 citation· ⚡1
This framework surveys manipulation, navigation, locomotion, autonomous driving, and general embodied learning, tracing technical progressions, clarifying capability requirements, and examining datasets, benchmarks, and evaluation protocols.