In-Context Policy Adaptation via Cross-Domain Skill Diffusion

• URL: http://arxiv.org/abs/2509.04535v1
• Date: Thu, 04 Sep 2025 06:55:38 GMT
• Title: In-Context Policy Adaptation via Cross-Domain Skill Diffusion
• Authors: Minjong Yoo, Woo Kyung Kim, Honguk Woo,
• Abstract summary: In this work, we present an in-context policy adaptation framework designed for long-horizon multi-task environments.<n>The framework enables rapid adaptation of skill-based reinforcement learning policies to diverse target domains.<n>We show that our framework achieves superior policy adaptation performance under limited target domain data conditions.
• Score: 37.727612185480986
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: In this work, we present an in-context policy adaptation (ICPAD) framework designed for long-horizon multi-task environments, exploring diffusion-based skill learning techniques in cross-domain settings. The framework enables rapid adaptation of skill-based reinforcement learning policies to diverse target domains, especially under stringent constraints on no model updates and only limited target domain data. Specifically, the framework employs a cross-domain skill diffusion scheme, where domain-agnostic prototype skills and a domain-grounded skill adapter are learned jointly and effectively from an offline dataset through cross-domain consistent diffusion processes. The prototype skills act as primitives for common behavior representations of long-horizon policies, serving as a lingua franca to bridge different domains. Furthermore, to enhance the in-context adaptation performance, we develop a dynamic domain prompting scheme that guides the diffusion-based skill adapter toward better alignment with the target domain. Through experiments with robotic manipulation in Metaworld and autonomous driving in CARLA, we show that our $\oursol$ framework achieves superior policy adaptation performance under limited target domain data conditions for various cross-domain configurations including differences in environment dynamics, agent embodiment, and task horizon.

Related papers

• Self-Paced Collaborative and Adversarial Network for Unsupervised Domain Adaptation [74.27130400558013]
  This paper proposes a new unsupervised domain adaptation approach called Collaborative and Adversarial Network (CAN)<n>CAN uses the domain-collaborative and domain-adversarial learning strategy for training the neural network.<n>To further enhance the discriminability in the target domain, we propose Self-Paced CAN (SPCAN)
  arXiv  Detail & Related papers  (2025-06-24T02:58:37Z)
• xTED: Cross-Domain Adaptation via Diffusion-Based Trajectory Editing [21.37585797507323]
  Cross-domain policy transfer methods mostly aim at learning domain correspondences or corrections to facilitate policy learning.<n>We propose the Cross-Domain Trajectory EDiting framework that employs a specially designed diffusion model for cross-domain trajectory adaptation.<n>Our proposed model architecture effectively captures the intricate dependencies among states, actions, and rewards, as well as the dynamics patterns within target data.
  arXiv  Detail & Related papers  (2024-09-13T10:07:28Z)
• Cross-Domain Policy Transfer by Representation Alignment via Multi-Domain Behavioral Cloning [13.674493608667627]
  We present a simple approach for cross-domain policy transfer that learns a shared latent representation across domains and a common abstract policy on top of it. Our approach leverages multi-domain behavioral cloning on unaligned trajectories of proxy tasks and employs maximum mean discrepancy (MMD) as a regularization term to encourage cross-domain alignment.
  arXiv  Detail & Related papers  (2024-07-24T00:13:00Z)
• Adapt in Contexts: Retrieval-Augmented Domain Adaptation via In-Context Learning [48.22913073217633]
  Large language models (LLMs) have showcased their capability with few-shot inference known as in-context learning. In this paper, we study the UDA problem under an in-context learning setting to adapt language models from the source domain to the target domain without any target labels. We devise different prompting and training strategies, accounting for different LM architectures to learn the target distribution via language modeling.
  arXiv  Detail & Related papers  (2023-11-20T06:06:20Z)
• Cross-Domain Policy Adaptation via Value-Guided Data Filtering [57.62692881606099]
  Generalizing policies across different domains with dynamics mismatch poses a significant challenge in reinforcement learning. We present the Value-Guided Data Filtering (VGDF) algorithm, which selectively shares transitions from the source domain based on the proximity of paired value targets.
  arXiv  Detail & Related papers  (2023-05-28T04:08:40Z)
• AFAN: Augmented Feature Alignment Network for Cross-Domain Object Detection [90.18752912204778]
  Unsupervised domain adaptation for object detection is a challenging problem with many real-world applications. We propose a novel augmented feature alignment network (AFAN) which integrates intermediate domain image generation and domain-adversarial training. Our approach significantly outperforms the state-of-the-art methods on standard benchmarks for both similar and dissimilar domain adaptations.
  arXiv  Detail & Related papers  (2021-06-10T05:01:20Z)
• Unsupervised Cross-domain Image Classification by Distance Metric Guided Feature Alignment [11.74643883335152]
  Unsupervised domain adaptation is a promising avenue which transfers knowledge from a source domain to a target domain. We propose distance metric guided feature alignment (MetFA) to extract discriminative as well as domain-invariant features on both source and target domains. Our model integrates class distribution alignment to transfer semantic knowledge from a source domain to a target domain.
  arXiv  Detail & Related papers  (2020-08-19T13:36:57Z)
• Domain Conditioned Adaptation Network [90.63261870610211]
  We propose a Domain Conditioned Adaptation Network (DCAN) to excite distinct convolutional channels with a domain conditioned channel attention mechanism. This is the first work to explore the domain-wise convolutional channel activation for deep DA networks.
  arXiv  Detail & Related papers  (2020-05-14T04:23:24Z)
• Improve Unsupervised Domain Adaptation with Mixup Training [18.329571222689562]
  We study the problem of utilizing a relevant source domain with abundant labels to build predictive modeling for an unannotated target domain. Recent work observe that the popular adversarial approach of learning domain-invariant features is insufficient to achieve desirable target domain performance. We propose to enforce training constraints across domains using mixup formulation to directly address the generalization performance for target data.
  arXiv  Detail & Related papers  (2020-01-03T01:21:27Z)

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.