# GeoLAM: Learning Geometry-Grounded Latent Actions from Unlabeled Human Videos

arXiv: https://arxiv.org/abs/2609.17099 · HTML: https://mlsanity.com/paper/2609.17099/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 39 of its 40 references that are in the map.
- First posted on arXiv 15 Sep 2026.

## Builds on (most cited first, 30 of 39)

- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [Wan: Open and Advanced Large-Scale Video Generative Models](https://mlsanity.com/paper/2503.20314/) (2025)
- [π0.5: a Vision-Language-Action Model with Open-World Generalization](https://mlsanity.com/paper/2504.16054/) (2025)
- [Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware](https://mlsanity.com/paper/2304.13705/) (2023)
- [Ego4D: Around the World in 3,000 Hours of Egocentric Video](https://mlsanity.com/paper/2110.07058/) (2021)
- [LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning](https://mlsanity.com/paper/2306.03310/) (2023)
- [The "something something" video database for learning and evaluating visual common sense](https://mlsanity.com/paper/1706.04261/) (2017)
- [Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success](https://mlsanity.com/paper/2502.19645/) (2025)
- [Depth Anything 3: Recovering the Visual Space from Any Views](https://mlsanity.com/paper/2511.10647/) (2025)
- [Rescaling Egocentric Vision](https://mlsanity.com/paper/2006.13256/) (2020)
- [RoboTwin 2.0: A Scalable Data Generator and Benchmark with Strong Domain Randomization for Robust Bimanual Robotic Manipulation](https://mlsanity.com/paper/2506.18088/) (2025)
- [UniVLA: Learning to Act Anywhere with Task-centric Latent Actions](https://mlsanity.com/paper/2505.06111/) (2025)
- [Fast-WAM: Do World Action Models Need Test-time Future Imagination?](https://mlsanity.com/paper/2603.16666/) (2026)
- [Latent Action Pretraining from Videos](https://mlsanity.com/paper/2410.11758/) (2024)
- [Causal World Modeling for Robot Control](https://mlsanity.com/paper/2601.21998/) (2026)
- [X-VLA: Soft-Prompted Transformer as Scalable Cross-Embodiment Vision-Language-Action Model](https://mlsanity.com/paper/2510.10274/) (2025)
- [Motus: A Unified Latent Action World Model](https://mlsanity.com/paper/2512.13030/) (2025)
- [EgoDex: Learning Dexterous Manipulation from Large-Scale Egocentric Video](https://mlsanity.com/paper/2505.11709/) (2025)
- [HoloAssist: an Egocentric Human Interaction Dataset for Interactive AI Assistants in the Real World](https://mlsanity.com/paper/2309.17024/) (2023)
- [ABot-M0: VLA Foundation Model for Robotic Manipulation with Action Manifold Learning](https://mlsanity.com/paper/2602.11236/) (2026)
- [HaWoR: World-Space Hand Motion Reconstruction from Egocentric Videos](https://mlsanity.com/paper/2501.02973/) (2025)
- [villa-X: Enhancing Latent Action Modeling in Vision-Language-Action Models](https://mlsanity.com/paper/2507.23682/) (2025)
- [EgoVerse: An Egocentric Human Dataset for Robot Learning from Around the World](https://mlsanity.com/paper/2604.07607/) (2026)
- [Efficiently Reconstructing Dynamic Scenes One D4RT at a Time](https://mlsanity.com/paper/2512.08924/) (2025)
- [Xiaomi-Robotics-1: Scaling Vision-Language-Action Models with over 100K Hours of Real-World Trajectories](https://mlsanity.com/paper/2607.15330/) (2026)
- [OA-WAM: Object-Addressable World Action Model for Robust Robot Manipulation](https://mlsanity.com/paper/2605.06481/) (2026)
- [EgoLive: A Large-Scale Egocentric Dataset from Real-World Human Tasks](https://mlsanity.com/paper/2604.23570/) (2026)
- [LAOF: Robust Latent Action Learning with Optical Flow Constraints](https://mlsanity.com/paper/2511.16407/) (2025)
- [Repurposing Geometric Foundation Models for Multi-view Diffusion](https://mlsanity.com/paper/2603.22275/) (2026)

## Abstract (arXiv)

Human videos provide rich manipulation experience, but extracting action representations that preserve useful motion remains challenging. Visual reconstruction alone can entangle manipulation-related motion with appearance changes and camera movement. We present GeoLAM, a framework for learning geometry-grounded latent actions from action-free human videos. GeoLAM combines future-frame reconstruction through a frozen geometric feature hierarchy with motion supervision from a training-only 4D geometry teacher. The geometric representation provides a structural prior, while the teacher's predictions yield spatially pooled targets capturing 3D displacement, residual image-plane motion, and surface-orientation changes. Visibility and confidence weighting reduces the contribution of unreliable estimates, encouraging continuous latent actions to retain geometric motion without explicit hand-pose or hand-trajectory annotations. After video pretraining without action labels, the learned representation provides transition targets for a world-action model trained on action-labeled robot demonstrations. The model jointly denoises latent actions and executable action chunks, with future-video prediction used only as an auxiliary training task. Deployment therefore requires neither the geometry teacher nor future-video generation. Evaluations on a latent-action benchmark and robotic manipulation tasks demonstrate the strong performance of GeoLAM.