# GIFT: Guided Intermediate Feature Training via Action-Oriented Structural Supervision for Robotic Manipulation arXiv: https://arxiv.org/abs/2609.04193 · HTML: https://mlsanity.com/paper/2609.04193/ ## Key facts - Cited by 4 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 33 of its 69 references that are in the map. - First posted on arXiv 3 Sep 2026. ## Placed in - World models / [Embodied and robot world models](https://mlsanity.com/wm/embodied-and-robot-world-models/) - Multimodal LLMs / vision-language models / [Embodied and 3D vision-language models](https://mlsanity.com/mllm/embodied-and-3d-vision-language-models/) ## Builds on (most cited first, 30 of 33) - [GR00T N1: An Open Foundation Model for Generalist Humanoid Robots](https://mlsanity.com/paper/2503.14734/) (2025) - [SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics](https://mlsanity.com/paper/2506.01844/) (2025) - [WorldVLA: Towards Autoregressive Action World Model](https://mlsanity.com/paper/2506.21539/) (2025) - [Fast-WAM: Do World Action Models Need Test-time Future Imagination?](https://mlsanity.com/paper/2603.16666/) (2026) - [LIBERO-Plus: In-depth Robustness Analysis of Vision-Language-Action Models](https://mlsanity.com/paper/2510.13626/) (2025) - [Genie Envisioner: A Unified World Foundation Platform for Robotic Manipulation](https://mlsanity.com/paper/2508.05635/) (2025) - [Interactive Post-Training for Vision-Language-Action Models](https://mlsanity.com/paper/2505.17016/) (2025) - [Spatial Forcing: Implicit Spatial Representation Alignment for Vision-language-action Model](https://mlsanity.com/paper/2510.12276/) (2025) - [NORA: A Small Open-Sourced Generalist Vision Language Action Model for Embodied Tasks](https://mlsanity.com/paper/2504.19854/) (2025) - [StarVLA: A Lego-like Codebase for Vision-Language-Action Model Developing](https://mlsanity.com/paper/2604.05014/) (2026) - [ABot-M0: VLA Foundation Model for Robotic Manipulation with Action Manifold Learning](https://mlsanity.com/paper/2602.11236/) (2026) - [Video Generators are Robot Policies](https://mlsanity.com/paper/2508.00795/) (2025) - [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) - [GeoVLA: Empowering 3D Representations in Vision-Language-Action Models](https://mlsanity.com/paper/2508.09071/) (2025) - [World Action Models: The Next Frontier in Embodied AI](https://mlsanity.com/paper/2605.12090/) (2026) - [CLAM: Continuous Latent Action Models for Robot Learning from Unlabeled Demonstrations](https://mlsanity.com/paper/2505.04999/) (2025) - [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) - [RLDX-1 Technical Report](https://mlsanity.com/paper/2605.03269/) (2026) - [OA-WAM: Object-Addressable World Action Model for Robust Robot Manipulation](https://mlsanity.com/paper/2605.06481/) (2026) - [DIAL: Decoupling Intent and Action via Latent World Modeling for End-to-End VLA](https://mlsanity.com/paper/2603.29844/) (2026) - [HumanEgo: Zero-Shot Robot Learning from Minutes of Human Egocentric Videos](https://mlsanity.com/paper/2605.24934/) (2026) - [MaskWAM: Unifying Mask Prompting and Prediction for World-Action Models](https://mlsanity.com/paper/2606.13515/) (2026) - [TwinBrainVLA: Unleashing the Potential of Generalist VLMs for Embodied Tasks via Asymmetric Mixture-of-Transformers](https://mlsanity.com/paper/2601.14133/) (2026) - [AffordanceVLA: A Vision-Language-Action Model Empowering Action Generation through Affordance-Aware Understanding](https://mlsanity.com/paper/2606.06155/) (2026) - [GeoSem-WAM: Geometry- and Semantic-Aware World Action Models](https://mlsanity.com/paper/2606.03188/) (2026) - [Geometric Action Model for Robot Policy Learning](https://mlsanity.com/paper/2606.17046/) (2026) - [Flex-π: A Multi-Stream World-Action Model with Compute Flexibility](https://mlsanity.com/paper/2608.10860/) (2026) - [PokeVLA: Empowering Pocket-Sized Vision-Language-Action Model with Comprehensive World Knowledge Guidance](https://mlsanity.com/paper/2604.20834/) (2026) ## Built on by (most cited first, 4 of 4) - [SCULPT-VLA: Learning Structured Control through Staged Action Grounding](https://mlsanity.com/paper/2609.23275/) (2026) - [From World Models to World Action Models: Rethinking Next-State Prediction](https://mlsanity.com/paper/2609.34414/) (2026) - [GeoBridge-VLA: Geometry-Aware Residual Adaptation for Vision-Language-Action Models](https://mlsanity.com/paper/2610.05026/) (2026) - [OpenWAM: An Open Framework for Composable World-Action Models](https://mlsanity.com/paper/2610.07922/) (2026) ## Abstract (arXiv) Vision-language pre-training and predictive world modeling provide robot policies with rich semantic and dynamic visual features, but their native action and visual-prediction objectives may omit critical physical and task structure while retaining control-irrelevant visual redundancy. We call this mismatch between visual richness and control utility the action-sufficiency gap. We investigate whether this gap can be bridged by guiding intermediate features to preserve three control-relevant structure in robotic manipulation: geometry governing motion feasibility, affordance encoding instruction-relevant entities, and goals grounding instructions in task-relevant regions. To this end, we present GIFT (Guided Intermediate Feature Training), an architecture-flexible framework for learning intermediate features that translates these structures into training-time constraints through geometry alignment, affordance prediction, and goal-region reconstruction. We instantiate GIFT in a Vision-Language-Action (VLA) policy, a direct-action World-Action Model (WAM), and an inverse-dynamics WAM while retaining each model's action formulation. Under zero-shot transfer to LIBERO-Plus, GIFT-VLA, GIFT-WAM-Fast, and GIFT-WAM-IDM outperform StarVLA-OFT, Fast-WAM, and Fast-WAM-IDM by 4.6, 12.6, and 5.2 points, reaching 79.6%, 72.6%, and 87.8%, respectively. On RoboCasa, the three GIFT variants reach 61.4%, 83.6%, and 82.3%, outperforming their counterparts by 12.6, 9.0, and 8.4 points, respectively. Together, these results establish learning functionally structured intermediate features as a reusable principle across model-specific action formulations, with especially large gains on articulated-object tasks and high-precision real-world manipulation under unseen visual and spatial perturbations. Project page: https://openphoenix-team.github.io/GIFT-pages.