# WAM-TTT: Steering World-Action Models by Watching Human Play at Test Time arXiv: https://arxiv.org/abs/2607.06988 · HTML: https://mlsanity.com/paper/2607.06988/ ## Key facts - Cited by 8 papers in the map (data through 14 Oct 2026). - Placed in Embodied and robot world models, one of 11 directions in World models. - Builds on 38 of its 61 references that are in the map. - First posted on arXiv 8 Jul 2026. ## Placed in - World models / [Embodied and robot world models](https://mlsanity.com/wm/embodied-and-robot-world-models/) ## Builds on (most cited first, 30 of 38) - [π0.5: a Vision-Language-Action Model with Open-World Generalization](https://mlsanity.com/paper/2504.16054/) (2025) - [DINOv3](https://mlsanity.com/paper/2508.10104/) (2025) - [MediaPipe: A Framework for Building Perception Pipelines](https://mlsanity.com/paper/1906.08172/) (2019) - [World Action Models are Zero-shot Policies](https://mlsanity.com/paper/2602.15922/) (2026) - [Fast-WAM: Do World Action Models Need Test-time Future Imagination?](https://mlsanity.com/paper/2603.16666/) (2026) - [Titans: Learning to Memorize at Test Time](https://mlsanity.com/paper/2501.00663/) (2024) - [Causal World Modeling for Robot Control](https://mlsanity.com/paper/2601.21998/) (2026) - [Video Prediction Policy: A Generalist Robot Policy with Predictive Visual Representations](https://mlsanity.com/paper/2412.14803/) (2024) - [Unified World Models: Coupling Video and Action Diffusion for Pretraining on Large Robotic Datasets](https://mlsanity.com/paper/2504.02792/) (2025) - [MimicPlay: Long-Horizon Imitation Learning by Watching Human Play](https://mlsanity.com/paper/2302.12422/) (2023) - [EgoDex: Learning Dexterous Manipulation from Large-Scale Egocentric Video](https://mlsanity.com/paper/2505.11709/) (2025) - [EgoMimic: Scaling Imitation Learning via Egocentric Video](https://mlsanity.com/paper/2410.24221/) (2024) - [Cosmos 3: Omnimodal World Models for Physical AI](https://mlsanity.com/paper/2606.02800/) (2026) - [Gen2Act: Human Video Generation in Novel Scenarios enables Generalizable Robot Manipulation](https://mlsanity.com/paper/2409.16283/) (2024) - [GigaWorld-Policy: An Efficient Action-Centered World--Action Model](https://mlsanity.com/paper/2603.17240/) (2026) - [π0.7: a Steerable Generalist Robotic Foundation Model with Emergent Capabilities](https://mlsanity.com/paper/2604.15483/) (2026) - [ABot-M0: VLA Foundation Model for Robotic Manipulation with Action Manifold Learning](https://mlsanity.com/paper/2602.11236/) (2026) - [EgoScale: Scaling Dexterous Manipulation with Diverse Egocentric Human Data](https://mlsanity.com/paper/2602.16710/) (2026) - [DiT4DiT: Jointly Modeling Video Dynamics and Actions for Generalizable Robot Control](https://mlsanity.com/paper/2603.10448/) (2026) - [Being-H0.7: A Latent World-Action Model from Egocentric Videos](https://mlsanity.com/paper/2605.00078/) (2026) - [Unified 4D World Action Modeling from Video Priors with Asynchronous Denoising](https://mlsanity.com/paper/2604.26694/) (2026) - [WoW: Towards a World omniscient World model Through Embodied Interaction](https://mlsanity.com/paper/2509.22642/) (2025) - [Motubrain: An Advanced World Action Model for Robot Control](https://mlsanity.com/paper/2604.27792/) (2026) - [Vid2Robot: End-to-end Video-conditioned Policy Learning with Cross-Attention Transformers](https://mlsanity.com/paper/2403.12943/) (2024) - [LDA-1B: Scaling Latent Dynamics Action Model via Universal Embodied Data Ingestion](https://mlsanity.com/paper/2602.12215/) (2026) - [Zero-Shot Robot Manipulation from Passive Human Videos](https://mlsanity.com/paper/2302.02011/) (2023) - [AHA-WAM:Asynchronous Horizon-Adaptive World-Action Modeling with Observation-Guided Context Routing](https://mlsanity.com/paper/2606.09811/) (2026) - [ImageWAM: Do World Action Models Really Need Video Generation, or Just Image Editing?](https://mlsanity.com/paper/2606.19531/) (2026) - [ABot-M0.5: Unified Mobility-and-Manipulation World Action Model](https://mlsanity.com/paper/2607.00678/) (2026) - [EVA: An Embodied World Model for Future Video Anticipation](https://mlsanity.com/paper/2410.15461/) (2024) ## Built on by (most cited first, 8 of 8) - [Zero-WAM: In-Context World-Action Modeling from Human Videos for Open-Ended Task Generalization](https://mlsanity.com/paper/2608.26103/) (2026) - [StellaVLA: In-Context Structured Demonstration for Generalizable Vision-Language-Action Models](https://mlsanity.com/paper/2608.11671/) (2026) - [In-Context Learning for Robots: Methods and Applications](https://mlsanity.com/paper/2609.36012/) (2026) - [Long-WAM: Scaling the Context of World-Action Models](https://mlsanity.com/paper/2610.10528/) (2026) - [Learning to Use Imagination: Progress-Conditioned Future Utilization for World Action Models](https://mlsanity.com/paper/2609.06578/) (2026) - [JEPA-TTT: Persistent Test-Time Training of Latent World Models for Planning under Dynamics Shifts](https://mlsanity.com/paper/2610.00722/) (2026) - [Juno: Taming Predictive Latents for Vision-Language-Action Models](https://mlsanity.com/paper/2610.09940/) (2026) - [World-T3: Test-Time Training as Persistent Memory for Real-Time Video World Models](https://mlsanity.com/paper/2610.17434/) (2026) ## Abstract (arXiv) Steering robot foundation models (RFMs) toward new task variants or user-preferred behaviors remains challenging, often requiring additional robot demonstrations, task-specific fine-tuning, or long-context conditioning. We present WAM-TTT, a test-time training framework for steering world action models from raw human videos. Rather than treating human videos as trajectories to imitate, WAM-TTT absorbs them into a lightweight adaptive memory inside a frozen WAM through self-supervised video prediction. To make this memory useful for control, we introduce a meta-training stage that aligns human demonstrations with robot behaviors using paired human-robot data and a key--value memory reconstruction objective. At test time, only unlabeled human videos are required to adapt the memory, while the pretrained WAM remains frozen. This enables efficient and reusable steering without robot actions, human-side annotations, or task-specific fine-tuning, while preserving the generalization ability of the foundation model. Extensive experiments show that WAM-TTT consistently outperforms in-context human-video conditioning baselines across diverse manipulation tasks and generalization settings.