G$^2$WAM: World Action Models for Efficient and Geometry Grounded Image-Goal Navigation
College of Intelligence Science and Technology, National University of Defense Technology(国防科技大学智能科学学院);School of Future Technology, National University of Defense Technology(国防科技大学未来技术学院)
Ke Gui, Chenghao Shi, Ziyue Wang, Longwen Chen, Xieyuanli Chen, Huimin Lu
World models capture environment dynamics and support counterfactual reasoning for planning. In image-goal navigation, where the target often lies beyond the current field of view, the agent must not only model environmental dynamics but also maintain a persistent spatial memory of the scene. Existing approaches either introduce spatial context by fine-tuning pretrained models on the target task, which underexploits their spatial priors, or rely on external 3D reconstruction, which is computationally expensive for real-world deployment. In this paper, we propose the Geometry Grounded World Action Model (G$^2$WAM) for image-goal navigation. G$^2$WAM encodes spatial cues through a learned scene-coordinate localization module that anchors the current, goal, and history frames in a common reference frame, yielding a stable and consistent spatial context that enables efficient navigation without explicit 3D reconstruction. Following the world-action-model paradigm, G$^2$WAM jointly predicts future states and actions, eliminating the need for costly cross-entropy method (CEM) style planning. In addition, G$^2$WAM operates entirely in latent space, avoiding the computational burden of pixel-level reconstruction and substantially reducing inference cost. Across four trajectory prediction benchmarks, G$^2$WAM reduces average ATE and RPE by 40.1% and 35.2%, respectively. In closed-loop navigation experiments, it improves the success rate from 54.0% to 63.0% and the success weighted by path length (SPL) from 36.4% to 59.2%, while maintaining high efficiency.
WNM-3D: A World Navigation Model with 3D Scene Conditioning for Closed-Loop VLN
Institute of Artificial Intelligence, China Telecom(中国电信人工智能研究院);Zhejiang University(浙江大学);Tongji University(同济大学);Shanghai Jiao Tong University(上海交通大学)
MoMa-Kitchen: A 100K+ Benchmark for Affordance-Grounded Last-Mile Navigation in Mobile Manipulation
Fudan University(复旦大学);Shanghai AI Laboratory(上海人工智能实验室);University of Science and Technology of China(中国科学技术大学);Northwestern Polytechnical University(西北工业大学);TeleAI, China Telecom Corp Ltd(中国电信 TeleAI 公司)
WorldMAP: Bootstrapping Vision-Language Navigation Trajectory Prediction with Generative World Models
Harbin Institute of Technology(哈尔滨工业大学);Shandong University(山东大学);Tsinghua University(清华大学);University of Science and Technology of China(中国科学技术大学);Zhongguancun Academy(中关村学院)
Korea University(韩国大学);University of Michigan(密歇根大学);Samsung AI Center, DS Division(三星人工智能中心,DS部门);Korea Advanced Institute of Science & Technology(韩国先进科学技术研究所)
Grounded world models in biological organisms and future embodied AI
Institute of Cognitive Sciences and Technologies, National Research Council(国家研究委员会认知科学与技术研究所);Center for Mind and Brain Science (CIMeC), University of Trento(特伦托大学心智与脑科学中心(CIMeC));University of Montreal(蒙特利尔大学)
VISOR: VIsual Spatial Object Reasoning for Language-driven Object Navigation
Polytechnic of Turin(都灵理工大学);Simon Fraser University(Simon Fraser大学);University of Verona(威尼斯大学);University of Reykjavik(雷克雅未克大学);University of California, Irvine(加州大学尔湾分校)
GA-VLN: Geometry-Aware BEV Representation for Efficient Vision-Language Navigation
State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences(人工智能安全国家重点实验室,计算技术研究所,中国科学院);University of Chinese Academy of Sciences(中国科学院大学);Robbyant;School of Computing, National University of Singapore(新加坡国立大学计算机学院);The Hong Kong University of Science and Technology(香港科技大学)