In-Context Learning for Robots: Methods and Applications

Haojian Huang, Zexi Li, Junhao Guo, Yehang Zhang, Wenxuan Peng, Bohan Zhou, Weilin Ruan, Leyi Wu, Chenxu Wang, Jianchong Su, Binghui Xie, Wosong Chen, Yingjie Xu, Tianhao Zhou, Suzeyu Chen, Pukun Zhao, Jiaqi He, Xinyi Li, Runze Li, Peiran Dong, Shaoxiang Dang, Jing Huang, Yingbing Chen, Yifan Chang, Tianyi Zhang, Shiyuan Deng, Haozhi Wang, Yangkai Wei, Wenqian Li, Han Yang, Kaiwen Zhou, Huaping Liu, James Cheng, Rui Shao, Donglin Wang, Yaochu Jin, Jianye Hao, Ying-Cong Chen, Yinchuan Li · N/A · 2026

Summary

General-purpose robots must infer what a new task requires and translate that understanding into appropriate physical action.

Abstract Summary

General-purpose robots must infer what a new task requires and translate that understanding into appropriate physical action. In-context learning (ICL) for robots supports this process by using demonstrations and interaction to direct existing competence with neural parameters held fixed during deployment. We organize this literature review around the interfaces connecting contextual evidence to execution, distinguishing four families: context-conditioned policies, geometric demonstration transfer, world-model-based control, and skill- and agent-based execution. Comparing these interfaces clarifies their transfer assumptions and the roles of training, correspondence, and memory in making context useful. Across manipulation and navigation, we examine how these mechanisms preserve taught requirements as objects, environments, and execution conditions change. This analysis links method design to evaluation practices that distinguish responsiveness to teaching, physical transfer, and benefits from retained experience. The resulting agenda connects compositional task acquisition and faithful transfer with physical recursive self-improvement, in which experience improves the ability to learn subsequent tasks.

Key Points

General-purpose robots must infer what a new task requires and translate that understanding into appropriate physical action.
In-context learning (ICL) for robots supports this process by using demonstrations and interaction to direct existing competence with neural parameters held fixed during deployment.
We organize this literature review around the interfaces connecting contextual evidence to execution, distinguishing four families: context-conditioned policies, geometric demonstration transfer, world-model-based control, and skill- and agent-based execution.
Comparing these interfaces clarifies their transfer assumptions and the roles of training, correspondence, and memory in making context useful.
Across manipulation and navigation, we examine how these mechanisms preserve taught requirements as objects, environments, and execution conditions change.

Foundations: Why Do Robots Need Context?

A packing robot may already grasp every part yet need a demonstration to establish their order, history to identify completed placements, and contact evidence to accommodate a tighter fit. The same visible scene can therefore require different actions. Useful context resolves this decision-relevant ambiguity through actions the robot can already execute.

This distinction motivates the learning horizons, adaptation regimes, and four control interfaces below.

Foundations: Why Do Robots Need Context? — In-Context Learning for Robots: Methods and Applications — Figure 3 : Technical foundations for six learning horizons. S1–S2 establish programmed and learned competence. S3 and S4 adapt execution through interaction history and supplied teaching, respectively, with fixed neural parameters. The S5 and S6 columns collect mechanisms supporting improvements in subsequent learning and knowledge exchange across embodiments, including reusable procedures, shared representations, and policy transfer.

Sources and Demonstrations

Figures are reproduced from Huang et al. See the full paper for experimental details and the project page, when the paper links one, for demonstrations and videos.