InternW0-$Δ$: A World Action Model Bridging Predictive Dynamics and Actions with 20K+ Hours of Open Data

Xingyu Miao; Zizun Li; Baole Fang; Kaiwen Song; Tenghui Wang; Hanxue Zhang; Yating Wang; Xudong Li; Yuping He; Xueyuan Wei; Chao Gao; Xijie Yang; Yingxiang Xu; Kerui Ren; Wenqi Guo; Jianjun Zhou; Xinzhe Wang; Weiguang Zhao; Ni Yang; Zetao Cai; Yufei Xue; Hengjie Li; Zeyu He; Yuanzhen Zhou; Rong Fu; Jianyang Zhang; Siwei Cui; Fuxian Huang; Yunsong Zhou; Xing Gao; Yifei Yao; Qiaojun Yu; Kailin Li; Ming Zhou; Mu Huang; Xinyue Li; Wenze Cui; Bingqi Jiang; Xueyue Zhu; Junting Dong; Haoyu Guo; Tao Lu; Mulin Yu; Bowen Zhou; Bin Zhao; Tianfan Xue; Weinan Zhang; Chunhua Shen

arXiv · 2609.31394

InternW0-$Δ$: A World Action Model Bridging Predictive Dynamics and Actions with 20K+ Hours of Open Data

View original source ↗Read source document ↗Export metadata ↓

Abstract

World Action Models (WAMs) jointly model visual dynamics and action generation for generalist robot manipulation. A central challenge is to integrate priors from large-scale pretrained models---including visual dynamics, scene semantics, geometry, and motion---into a unified framework for robot action generation. We introduce InternW0-$Δ$, a unified WAM pretrained on a heterogeneous corpus that outperforms prior methods across simulation benchmarks and real-robot platforms. InternW0-$Δ$ combines pretrained visual dynamics, scene-level semantics, 4D geometric and motion priors, and action generation within a Mixture-of-Transformers (MoT) framework. A pretrained video expert and an action expert interact under semantic guidance from a frozen VLM, while a pretrained 4D foundation model injects geometric and motion priors through training-only distillation. We further introduce Causal Imprint, which learns future-relevant scene changes from training-only future supervision and provides predictive representations directly to the action expert without future-video rollout at inference. For large-scale joint training, we construct a heterogeneous corpus of robot demonstrations, UMI data, egocentric human demonstrations, and Ego2Robot data, curated and aligned under a common state-action representation. The resulting corpus contains over 20K hours of processed training data, to our knowledge the largest open-source corpus of its kind. We pretrain InternW0-$Δ$ on this corpus and demonstrate strong performance across simulation benchmarks and real-robot platforms. We will open source the training code, model weights, infrastructure, data-processing pipeline, and processed data where licenses permit. Project page: https://internrobotics.github.io/InternW0-Delta/

Explore related subjects

cs.RO cs.CV

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Xingyu Miao, Zizun Li, Baole Fang, Kaiwen Song, Tenghui Wang, Hanxue Zhang, Yating Wang, Xudong Li, Yuping He, Xueyuan Wei, Chao Gao, Xijie Yang, Yingxiang Xu, Kerui Ren, Wenqi Guo, Jianjun Zhou, Xinzhe Wang, Weiguang Zhao, Ni Yang, Zetao Cai, Yufei Xue, Hengjie Li, Zeyu He, Yuanzhen Zhou, Rong Fu, Jianyang Zhang, Siwei Cui, Fuxian Huang, Yunsong Zhou, Xing Gao, Yifei Yao, Qiaojun Yu, Kailin Li, Ming Zhou, Mu Huang, Xinyue Li, Wenze Cui, Bingqi Jiang, Xueyue Zhu, Junting Dong, Haoyu Guo, Tao Lu, Mulin Yu, Bowen Zhou, Bin Zhao, Tianfan Xue, Weinan Zhang, Chunhua Shen. 2026-09-25. InternW0-$Δ$: A World Action Model Bridging Predictive Dynamics and Actions with 20K+ Hours of Open Data. https://arxiv.org/abs/2609.31394

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Easier Said Than Done: Unpacking Intent-Behavior Gap in Jailbreaking LLM-based Robots

LLM-based robots use Large Language Models (LLMs) as planners to translate natural language instructions into policies such as grasp(), move_to(), and open_gripper(). Jailbreak attacks on these robots extend the threat from generating malicious content to executing harmful behaviors. However, we find that existing jailbreak attempts against LLM-based robots that produce malicious-looking policies (intent jailbreaks) often fail to induce harmful physical actions by robots (behavior jailbreaks), due to robot-specific constraints, such as logical errors and hallucinated control APIs. In this paper, we demystify the intent-behavior gap and investigate its root causes to inform effective defenses. Our measurement study finds that current LLM jailbreak methods overlook robot-specific syntax constraints (e.g., executable control APIs) and physical feasibility (e.g., ordering of policies and hardware/kinematic constraints). To bridge the gap, we introduce POEF (Policy Effective Jailbreak), an automated red-teaming framework that takes into account the robot-specific constraints during both the optimization and evaluation processes. Specifically, POEF employs the hidden-layer gradients from an unaligned LLM to guide the jailbreak prompt optimization and uses a multi-agent evaluator to assess the feasibility of the generated policies. Experiments on commercial robots, including the Unitree G1, the Franka robotic arm, and simulators, show that POEF achieves an 80% behavior jailbreak success rate and transfers across various LLMs. In addition, we propose two defense strategies that mitigate the behavior jailbreak risks. Our findings indicate an urgent need for stronger countermeasures before LLM-based robots are deployed at scale. The homepage is available at https://zjushine.github.io/poef.github.io/.

cs.RO↗

arXiv2026

SPIDER: Scalable Physics-Informed Dexterous Retargeting

Learning agile robotic policies requires large-scale demonstrations, but translating abundant human motion data to robots is bottlenecked by the embodiment gap and missing dynamic information. To bridge this gap, we propose Scalable Physics-Informed DExterous Retargeting (SPIDER), a physics-based retargeting framework to transform and augment kinematic-only human demonstrations into dynamically feasible robot trajectories at scale. Our key insight is that human demonstrations should provide global task structure and objective, while a sampling-based solver can be used to find the feasible solution given the physics constraints. As a general framework, SPIDER is an efficient physics-based retargeting method that can be applied to both humanoid whole-body loco-manipulation and dexterous manipulation across 9 humanoid/dexterous hand embodiments, 6 datasets and 3 simulators. By bypassing the need for policy optimization, it achieves state-of-the-art performance while being 10x faster than reinforcement learning (RL) baselines. Furthermore, SPIDER enables physics-based data augmentation, such as imposing external payloads, perturbations, and new contact patterns, to generate diverse data. We demonstrate that the retargeted motion can be executed on the robot directly open-loop or serve as feasible reference motion for efficient RL policy learning. Our pipeline enables large-scale generation of robot trajectories from human motion, and we release a dataset of 6,520 simulated demonstrations across four hands and 191 dataset-specific object identities to support future research.

cs.RO↗

arXiv2026

Learning Control Policies to Provably Satisfy Hard Affine Constraints for Black-Box Hybrid Dynamical Systems

Ensuring safety for black-box hybrid dynamical systems presents significant challenges due to their instantaneous state jumps and unknown explicit nonlinear dynamics. Existing solutions for strict safety constraint satisfaction, like control barrier functions (CBFs) and reachability analysis, rely on direct knowledge of the dynamics. Similarly, safe reinforcement learning (RL) approaches often rely on known system dynamics or merely discourage safety violations through reward shaping. In this work, we want to learn RL policies which provably satisfy affine state constraints in closed loop for black-box hybrid dynamical systems with affine reset maps. Our key insight is forcing the RL policy to be affine and repulsive near the constraint boundaries for the unknown nonlinear dynamics of the system, providing guarantees that the trajectories will not violate the constraint. We further account for constraint violation due to instantaneous state jumps that occur due to impacts or reset maps in the hybrid system by introducing a second repulsive affine region before the reset that prevents post-reset states from violating the constraint. We derive sufficient conditions under which these policies satisfy safety constraints in closed loop. We also compare our approach with state-of-the-art reward shaping and learned-CBF methods on hybrid dynamical systems like the constrained pendulum and paddle juggler environments. In both scenarios, we show that our methodology learns higher quality policies while always satisfying the safety constraints.

cs.RO↗