GigaWorld-Policy-0.5: A Faster and Stronger WAM Empowered by AutoResearch

Hugging Face Daily Papers Papers
world-action-models robot-control transformer-architecture efficient-inference auto-research pretraining mixture-of-experts

Summary

GigaWorld-Policy-0.5 is an enhanced World Action Model for robot control that improves training and inference efficiency through a Mixed Action-Conditioned World Modeling strategy and a Mixture-of-Transformers architecture, achieving 85ms latency on a local RTX 4090.

World Action Models (WAMs) improve robot policy learning by jointly modeling actions and future visual observations, using future scene evolution as dense supervision for physically grounded action generation. However, a common design in existing WAMs is to explicitly generate future videos at inference time, incurring substantial computational overhead and hindering real-time closed-loop deployment. GigaWorld-Policy addresses this issue with an action-centered formulation, where future visual dynamics are used during training while action-only decoding is used at inference time. Building upon this framework, we present GigaWorld-Policy-0.5, an enhanced action-centered WAM designed for more efficient robot control. During pretraining, GigaWorld-Policy-0.5 adopts a mixed Action-Conditioned World Modeling (AC-WM) and WAM training strategy. This strengthens the coupling between visual dynamics and robot actions and improves the transferability of action representations for downstream policy learning. For efficient inference, GigaWorld-Policy-0.5 introduces a Mixture-of-Transformers architecture that separates visual dynamics modeling and action generation into specialized experts, reducing active computation during action-only inference and achieving 85 ms inference latency on a local RTX 4090 setup. In addition, we employ an agent-based AutoResearch pipeline to systematically search training configurations, enabling more efficient identification of optimal experimental setups while reducing the time and manual intervention required for hyperparameter tuning. Experiments and ablations show that GigaWorld-Policy-0.5 preserves the training benefits of future visual dynamics while improving inference efficiency for robot control.
Original Article
View Cached Full Text

Cached at: 07/16/26, 05:42 AM

Paper page - GigaWorld-Policy-0.5: A Faster and Stronger WAM Empowered by AutoResearch

Source: https://huggingface.co/papers/2607.13960 Authors:

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

Abstract

WorldActionModels(WAMs)improverobotpolicylearningbyjointlymodelingactionsandfuturevisualobservations,usingfuturesceneevolutionasdensesupervisionforphysicallygroundedactiongeneration.However,acommondesigninexistingWAMsistoexplicitlygeneratefuturevideosatinferencetime,incurringsubstantialcomputationaloverheadandhinderingreal-timeclosed-loopdeployment.GigaWorld-Policyaddressesthisissuewithanaction-centeredformulation,wherefuturevisualdynamicsareusedduringtrainingwhileaction-onlydecodingisusedatinferencetime.Buildinguponthisframework,wepresentGigaWorld-Policy-0.5,anenhancedaction-centeredWAMdesignedformoreefficientrobotcontrol.Duringpretraining,GigaWorld-Policy-0.5adoptsamixedAction-ConditionedWorldModeling(AC-WM)andWAMtrainingstrategy.Thisstrengthensthecouplingbetweenvisualdynamicsandrobotactionsandimprovesthetransferabilityofactionrepresentationsfordownstreampolicylearning.Forefficientinference,GigaWorld-Policy-0.5introducesaMixture-of-Transformersarchitecturethatseparatesvisualdynamicsmodelingandactiongenerationintospecializedexperts,reducingactivecomputationduringaction-onlyinferenceandachieving85msinferencelatencyonalocalRTX4090setup.Inaddition,weemployanagent-basedAutoResearchpipelinetosystematicallysearchtrainingconfigurations,enablingmoreefficientidentificationofoptimalexperimentalsetupswhilereducingthetimeandmanualinterventionrequiredforhyperparametertuning.ExperimentsandablationsshowthatGigaWorld-Policy-0.5preservesthetrainingbenefitsoffuturevisualdynamicswhileimprovinginferenceefficiencyforrobotcontrol.

View arXiv pageView PDFProject pageAdd to collection

Get this paper in your agent:

hf papers read 2607\.13960

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper1

#### open-gigaai/Giga-World-Policy-0.5 Updatedabout 3 hours ago • 4

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2607.13960 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2607.13960 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

AHA-WAM:Asynchronous Horizon-Adaptive World-Action Modeling with Observation-Guided Context Routing

Hugging Face Daily Papers

AHA-WAM is an asynchronous world-action model that uses dual Diffusion Transformers to decouple world prediction from action execution, achieving efficient long-horizon planning and real-time control. It achieves state-of-the-art performance on robotic manipulation tasks with up to 92.8% success on RoboTwin and 78.3% on real-world tasks, while reaching 24.17 Hz closed-loop control.