1:04EgoWAM: World Action Models Beyond Pixels with In-the-Wild Egocentric Human Data@BaoyuLi6 · 101 views · 2026-07-08World-action modelsRobot ManipulationEgocentric Video
0:38Unitree UnifoLM-X2-1.0: Real-Time World-Model-Driven Humanoid Combat@UnitreeRobotics · 92 views · 2026-09-07UnitreeUnifoLMWorld Models
0:22Riemann-1.0 trains a World Action Model on 200k hours of human video@techniahqrobot · 198 views · 2026-08-29World-action modelsEmbodied AIHuman Video
DECOWAM: Decoupled Whole-Body World-Action Model for Legged Mobile ManipulationDECOWAM adapts a frozen FastWAM video-action backbone to legged mobile manipulation via decoupled interfaces — an action-equivalent future bottleneck, adversarial base/arm factorization, and ego-motion-aware video conditioning — cutting Stage-2 trainable parameters 232x while leading real-robot deployment at 58.2% success.Siyuan Ma, Boshi Zhang, Yutian Zhang·Aug 20, 2026VLAWorld ModelsWorld-action modelsAug 20, 2026
0:28Dyna-2 World-Action Model by Dyna Robotics@TheHumanoidHub · 64 views · 2026-08-10WAMDyna-2DYNA Robotics
1:42Dyna-2: World-Action Model Pre-trained on One Million Hours of Human Video@DynaRobotics · 63 views · 2026-08-10Dyna-2Scaling LawDynaRobotics
SimWAM: A Simple World Action Model for End-to-End Autonomous DrivingWorld-Action Models (WAMs) improve end-to-end autonomous driving by transferring video dynamics priors to action prediction, but existing methods require costly future generation at inference. We present SimWAM, a simple yet effective WAM that uses video generation purely as a training signal. It co-trains a pretrained video expert and a lightweight action expert with joint flow matching. An isolated attention mask keeps action prediction independent of future frames, allowing the video branch to be discarded after training and leaving a self-contained planner that directly predicts trajectories. Since the two experts share no parameters and interact only through a unified attention interface, the video backbone could be replaced and the action expert scaled independently without modifying the learning objective or inference pipeline. We further apply reinforcement learning to optimize a compositional driving reward beyond trajectory imitation. Our SimWAM achieves $91.5$ PDMS on NAVSIM, surpasses state-of-the-art WAM-based planners with substantially lower latency, and transfers zero-shot to nuScenes. These results position SimWAM as a simple yet solid baseline that could readily benefit from advances in video generation for efficient autonomous driving. The code and model weights are available at https://github.com/H-EmbodVis/SimWAM/Zongchuang Zhao, Xin Zhou, Tianyang Xu·Aug 7, 2026NAVSIMSimWAMWorld-action modelsAug 7, 2026
ω-0: A Latent Predictive World Action Model for Concurrent Humanoid Loco-Manipulationω-0 is a latent predictive whole-body world-action model for real-world humanoid concurrent loco-manipulation: given a language instruction, visual observation, and proprioceptive state, it directly predicts controller-compatible whole-body action latents, coupling compact future-observation embedding prediction with diffusion-based action generation. A 40+ hour real-world dataset ω-HOME is collected; a single model outperforms IL, VLA, humanoid, and WAM baselines on 11 household tasks.Zhe Li, Zhenzhe Zhang, Yangyang Wei·Aug 6, 2026World-action modelsHumanoidMobile ManipulationAug 6, 2026
Robust-WAM: Bridging Generative Pretraining and Semantic Foresight in World-Action ModelsMainstream World-Action Models (WAMs) adapt pretrained video generation models (VGMs) for robot control, transferring their learned dynamics prior for action prediction. These VGMs are typically trained in a variational autoencoder (VAE) latent space. However, the VAE latent space is optimized for pixel reconstruction, which rewards fine appearance detail and leaves the action prediction fragile under visual shifts. Recent works build WAMs in semantic latent space, which are more robust to appearance shifts. However, these models cannot leverage the large-scale VGM pretraining that exists only in VAE space. To overcome this dilemma, we propose Robust-WAM, a general post-training method for video-generation-based WAMs that preserves the VAE-based generative path and adds a lightweight semantic foresight alignment objective on the action stream. This retains the large-scale VGM pretraining while grounding actions in appearance-invariant dynamics that stay reliable under illumination shifts and other visual out-of-distribution conditions. Specifically, we employ learnable query tokens to bring future-scene semantics into the action stream by aligning their output hidden states with the semantic foresight of future ground-truth frames. To establish the temporal correspondence between each query and the future step it describes, we give it the positional encoding of the matching action tokens. Experiments on out-of-distribution generalization simulation benchmarks and a real-robot setup show that our Robust-WAM consistently improves the success rates of multiple WAM baselines without sacrificing in-distribution performance.Haodong Yan, Junfeng Li, Junjie He·Aug 6, 2026VAERobust-WAMWorld ModelsAug 6, 2026
1:04
