Abstract
We study robotic in-context learning (ICL), an emerging paradigm that enables robots to infer and execute tasks from visual demonstrations. Despite its growing promise, the problem itself remains under-defined: a visual demonstration simultaneously conveys action trajectories, object semantics, manipulation affordances, spatial relations, and task goals, making it unclear what information the robot is actually expected to follow. In this work, we first provide a clear problem definition of robot ICL that explicitly defines its learning target and resolves this fundamental prompt ambiguity. Building on this definition, we develop a minimalist and reproducible ICL framework (SimpleICL) with a visual prompt encoder and a low-cost data collection protocol. Without massive pre-training or specialized data infrastructure, our framework achieves strong performance in both simulation and real-world environments. Extensive experiments further reveal several key properties of robot ICL, including semantic discrimination and task-relevant disentanglement. We will fully open-source our data and training pipeline to facilitate systematic and reproducible research on robot ICL. The project page can be found at https://simpleicl.github.io/simpleicl.
Keywords
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Li, M., Han, M., Zhao, W., Wang, H., Liu, X., Shang, S., Zhou, J., Sun, M., Pan, H., Xu, M., Liu, Y., Fan, L., & Zhang, Z. (2026). In-context Robot Learning Made Simple: A Democratized Recipe for Manipulation Tasks. https://omanscience.com/en/articles/in-context-robot-learning-made-simple-a-democratized-recipe-for-manipulation-tasks
MLA 9
Li, Minxing, et al. "In-context Robot Learning Made Simple: A Democratized Recipe for Manipulation Tasks." https://omanscience.com/en/articles/in-context-robot-learning-made-simple-a-democratized-recipe-for-manipulation-tasks.
Chicago (author–date)
Li, Minxing, Minghao Han, Weizhi Zhao, Hanwen Wang, Xiangshuo Liu, Shuyao Shang, Jingxiang Zhou, Mingchao Sun, Hongyu Pan, Mu Xu, Yu Liu, Lue Fan, and Zhaoxiang Zhang. 2026. "In-context Robot Learning Made Simple: A Democratized Recipe for Manipulation Tasks." https://omanscience.com/en/articles/in-context-robot-learning-made-simple-a-democratized-recipe-for-manipulation-tasks.
Harvard
Li, M., Han, M., Zhao, W., Wang, H., Liu, X., Shang, S., Zhou, J., Sun, M., Pan, H., Xu, M., Liu, Y., Fan, L. and Zhang, Z. (2026) 'In-context Robot Learning Made Simple: A Democratized Recipe for Manipulation Tasks', Available at: https://omanscience.com/en/articles/in-context-robot-learning-made-simple-a-democratized-recipe-for-manipulation-tasks.
Vancouver
Li M, Han M, Zhao W, Wang H, Liu X, Shang S, et al. In-context Robot Learning Made Simple: A Democratized Recipe for Manipulation Tasks. https://omanscience.com/en/articles/in-context-robot-learning-made-simple-a-democratized-recipe-for-manipulation-tasks
IEEE
M. Li, M. Han, W. Zhao, H. Wang, X. Liu, S. Shang, J. Zhou, M. Sun, H. Pan, M. Xu, Y. Liu, L. Fan, and Z. Zhang, "In-context Robot Learning Made Simple: A Democratized Recipe for Manipulation Tasks," https://omanscience.com/en/articles/in-context-robot-learning-made-simple-a-democratized-recipe-for-manipulation-tasks.