Latent Policy Steering with Embodiment-Agnostic Pretrained World Models

• URL: http://arxiv.org/abs/2507.13340v1
• Date: Thu, 17 Jul 2025 17:57:57 GMT
• Title: Latent Policy Steering with Embodiment-Agnostic Pretrained World Models
• Authors: Yiqi Wang, Mrinal Verghese, Jeff Schneider,
• Abstract summary: We aim to reduce data collection efforts when learning visuomotor robot policies by leveraging existing or cost-effective data.<n>We use optic flow as an embodiment-agnostic action representation to train a World Model (WM) across multi-embodiment datasets.<n>We develop a method, Latent Policy Steering (LPS), to improve the output of a behavior-cloned policy by searching in the latent space of the WM for better action sequences.
• Score: 8.265847442964045
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Learning visuomotor policies via imitation has proven effective across a wide range of robotic domains. However, the performance of these policies is heavily dependent on the number of training demonstrations, which requires expensive data collection in the real world. In this work, we aim to reduce data collection efforts when learning visuomotor robot policies by leveraging existing or cost-effective data from a wide range of embodiments, such as public robot datasets and the datasets of humans playing with objects (human data from play). Our approach leverages two key insights. First, we use optic flow as an embodiment-agnostic action representation to train a World Model (WM) across multi-embodiment datasets, and finetune it on a small amount of robot data from the target embodiment. Second, we develop a method, Latent Policy Steering (LPS), to improve the output of a behavior-cloned policy by searching in the latent space of the WM for better action sequences. In real world experiments, we observe significant improvements in the performance of policies trained with a small amount of data (over 50% relative improvement with 30 demonstrations and over 20% relative improvement with 50 demonstrations) by combining the policy with a WM pretrained on two thousand episodes sampled from the existing Open X-embodiment dataset across different robots or a cost-effective human dataset from play.

Related papers

• Curating Demonstrations using Online Experience [52.59275477573012]
  We show that Demo-SCORE can effectively identify suboptimal demonstrations without manual curation.<n>Demo-SCORE achieves over 15-35% higher absolute success rate in the resulting policy compared to the base policy trained with all original demonstrations.
  arXiv  Detail & Related papers  (2025-03-05T17:58:16Z)
• STRAP: Robot Sub-Trajectory Retrieval for Augmented Policy Learning [8.860366821983211]
  STRAP is a technique for leveraging pre-trained vision foundation models and dynamic time warping to retrieve sub-sequences of trajectories from large training corpora in a robust fashion.<n>This work proposes STRAP, a technique for leveraging pre-trained vision foundation models and dynamic time warping to retrieve sub-sequences of trajectories from large training corpora in a robust fashion.
  arXiv  Detail & Related papers  (2024-12-19T18:54:06Z)
• Steering Your Generalists: Improving Robotic Foundation Models via Value Guidance [66.51390591688802]
  Value-Guided Policy Steering (V-GPS) is compatible with a wide range of different generalist policies, without needing to fine-tune or even access the weights of the policy.<n>We show that the same value function can improve the performance of five different state-of-the-art policies with different architectures.
  arXiv  Detail & Related papers  (2024-10-17T17:46:26Z)
• AdaDemo: Data-Efficient Demonstration Expansion for Generalist Robotic Agent [75.91274222142079]
  In this study, we aim to scale up demonstrations in a data-efficient way to facilitate the learning of generalist robotic agents. AdaDemo is a framework designed to improve multi-task policy learning by actively and continually expanding the demonstration dataset.
  arXiv  Detail & Related papers  (2024-04-11T01:59:29Z)
• Efficient Data Collection for Robotic Manipulation via Compositional Generalization [70.76782930312746]
  We show that policies can compose environmental factors from their data to succeed when encountering unseen factor combinations. We propose better in-domain data collection strategies that exploit composition. We provide videos at http://iliad.stanford.edu/robot-data-comp/.
  arXiv  Detail & Related papers  (2024-03-08T07:15:38Z)
• PoCo: Policy Composition from and for Heterogeneous Robot Learning [44.1315170137613]
  Current methods usually collect and pool all data from one domain to train a single policy.<n>We present a flexible approach, dubbed Policy Composition, to combine information across diverse modalities and domains.<n>Our method can use task-level composition for multi-task manipulation and be composed with analytic cost functions to adapt policy behaviors at inference time.
  arXiv  Detail & Related papers  (2024-02-04T14:51:49Z)
• Robot Fine-Tuning Made Easy: Pre-Training Rewards and Policies for Autonomous Real-World Reinforcement Learning [58.3994826169858]
  We introduce RoboFuME, a reset-free fine-tuning system for robotic reinforcement learning. Our insights are to utilize offline reinforcement learning techniques to ensure efficient online fine-tuning of a pre-trained policy. Our method can incorporate data from an existing robot dataset and improve on a target task within as little as 3 hours of autonomous real-world experience.
  arXiv  Detail & Related papers  (2023-10-23T17:50:08Z)
• Information Maximizing Curriculum: A Curriculum-Based Approach for Imitating Diverse Skills [14.685043874797742]
  We propose a curriculum-based approach that assigns a weight to each data point and encourages the model to specialize in the data it can represent. To cover all modes and thus, enable diverse behavior, we extend our approach to a mixture of experts (MoE) policy, where each mixture component selects its own subset of the training data for learning. A novel, maximum entropy-based objective is proposed to achieve full coverage of the dataset, thereby enabling the policy to encompass all modes within the data distribution.
  arXiv  Detail & Related papers  (2023-03-27T16:02:50Z)
• A Framework for Efficient Robotic Manipulation [79.10407063260473]
  We show that a single robotic arm can learn sparse-reward manipulation policies from pixels. We show that, given only 10 demonstrations, a single robotic arm can learn sparse-reward manipulation policies from pixels.
  arXiv  Detail & Related papers  (2020-12-14T22:18:39Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.