GreenWAM: A World Foundation Model for Robotics

Hugging Face Daily Papers Papers
robotics world-model foundation-model embodied-ai humanoid policy-learning

Summary

GreenWAM 提出一个机器人世界基础模型，通过共享的解剖学动作空间联合训练 20 个机器人家族，使其既能作为世界模型预测动作后果，又能作为策略控制不同形态的机器人，并在多个基准上取得领先表现。

A robot needs a world model that predicts how its actions will change the scene and a policy that chooses those actions. Models are usually built and evaluated for only one of these roles, and every robot records its actions in its own format. We ask whether one robot world model can serve both roles across different bodies, and what makes this work. We present GreenWAM, a world foundation model for robotics, together with the recipe that produces it: (1) a shared anatomical action space that describes 20 robot families, from bimanual arms to dexterous humanoids, in common joint coordinates; (2) joint training on these families in visual, action-conditioned and world-action modes; and (3) policies that split an action expert off the generator. As a world model, GreenWAM attains top-tier trajectory, depth and 3D accuracy on WorldArena 1.0, ahead of every world-action model and video generator in our comparison. As a policy, it scores among the top entries of the RoboTwin 2.0 leaderboard, reaches the highest zero-shot LIBERO-plus success among the compared methods and is competitive with dedicated vision-language-action and world-action policies on CALVIN and LIBERO. We further adapt GreenWAM to a real humanoid, where it succeeds more often than policies adapted from π_{0.5} and from Cosmos 3 Edge, and distill it into a single-step policy for real-time control. Controlled experiments yield three findings: robot specialization improves world prediction and control together; an action expert helps when split off at the start of adaptation; and keeping world-model modes during policy adaptation preserves robustness to scene changes.
Original Article
View Cached Full Text

Cached at: 10/14/26, 10:51 AM

Paper page - GreenWAM: A World Foundation Model for Robotics

Source: https://huggingface.co/papers/2610.18774

Abstract

Arobotneedsaworldmodelthatpredictshowitsactionswillchangethesceneandapolicythatchoosesthoseactions.Modelsareusuallybuiltandevaluatedforonlyoneoftheseroles,andeveryrobotrecordsitsactionsinitsownformat.Weaskwhetheronerobotworldmodelcanservebothrolesacrossdifferentbodies,andwhatmakesthiswork.WepresentGreenWAM,aworldfoundationmodelforrobotics,togetherwiththerecipethatproducesit:(1)asharedanatomicalactionspacethatdescribes20robotfamilies,frombimanualarmstodexteroushumanoids,incommonjointcoordinates;(2)jointtrainingonthesefamiliesinvisual,action-conditionedandworld-actionmodes;and(3)policiesthatsplitanactionexpertoffthegenerator.Asaworldmodel,GreenWAMattainstop-tiertrajectory,depthand3DaccuracyonWorldArena1.0,aheadofeveryworld-actionmodelandvideogeneratorinourcomparison.Asapolicy,itscoresamongthetopentriesoftheRoboTwin2.0leaderboard,reachesthehighestzero-shotLIBERO-plussuccessamongthecomparedmethodsandiscompetitivewithdedicatedvision-language-actionandworld-actionpoliciesonCALVINandLIBERO.WefurtheradaptGreenWAMtoarealhumanoid,whereitsucceedsmoreoftenthanpoliciesadaptedfromπ_{0.5}andfromCosmos3Edge,anddistillitintoasingle-steppolicyforreal-timecontrol.Controlledexperimentsyieldthreefindings:robotspecializationimprovesworldpredictionandcontroltogether;anactionexperthelpswhensplitoffatthestartofadaptation;andkeepingworld-modelmodesduringpolicyadaptationpreservesrobustnesstoscenechanges.

View arXiv pageView PDFProject pageGitHub1Add to collection

Models citing this paper1

#### SberRoboticsCenter/GreenWAM Robotics• Updatedabout 3 hours ago

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2610.18774 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2610.18774 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

AffordanceWAM: Affordance-Aware Joint World-Action Modeling for Robot Manipulation

Hugging Face Daily Papers

AffordanceWAM 提出一种感知可供性（affordance）的生成式世界-动作模型，利用人类视频中的可供性表征联合预测未来场景与机器人动作，无需人类动作标注即可实现人类到机器人的迁移，并在 RoboCasa、CALVIN 等基准上取得一致提升。

UniWAM: Unified World-Action Model

Hugging Face Daily Papers

UniWAM 提出了一种统一的架构，将物理推理器、世界生成器和动作预测器整合在一起，联合学习物理世界的语义理解、视觉生成与动作预测。该模型在多项评估中达到 SOTA，并揭示了统一人机协同训练的对数线性缩放规律。

@seclink: Robot World Models (New Dimension, 0 Deduplication = New Information) Core Projects: - Awesome-WAM (OpenMOSS): Comprehensive Paper List of World Action Models, including DreamDojo (General-Purpose Robot World Model Learned from Human Videos) - awe…

X AI KOLs Following

Introduces two projects related to robot world models: Awesome-WAM (OpenMOSS) includes papers such as World Action Models and DreamDojo; awesome-physical-ai curates a collection of papers on VLA models, world models, and embodied foundation models (including NVIDIA Cosmos Predict2.5).