cs.ROOct 30, 2026

Learning from Physical Experience:Context-Space Adaptation for Robotic Manipulation

Authors: Yihan Zeng, Yisu Li, Kaixin Huang, Marc WuDunn, Amelia Kovacs, Mikael Harseno, Kimmy Lin, Porter Hunley, 

Organizations: The Chinese University of Hong Kong · University of Electronic Science and Technology of China

Abstract

Recent multimodal foundation models exhibit strong perception, reasoning, and in-context learning capabilities, but translating these capabilities into physical action remains challenging. Through real-world experiments across diverse manipulation tasks, we find that ChatGPT-6-Astra, a state-of-the-art general-purpose reasoner, can serve as a physically actionable, context-responsive manipulation policy, demonstrating effective fine manipulation and spatial reasoning. Building on these findings, we introduce LEXP, a harness for learning from physical experience through context-space adaptation. LEXP combines task-level planning with an experience module that interprets execution evidence, extracts and retains lessons with applicability conditions, and reuses relevant lessons to generate contextual feedback for subsequent actions. Model weights and the action interface remain fixed throughout. Across 12 real-world manipulation tasks, LEXP achieves 23/3623/36 successful trials (63.9%), compared with 19/3619/36 (52.8%) with memory alone and 2/362/36 (5.6%) for vanilla Astra. On three tasks, bread transfer, sort trash and toiletries, and retrieve drawer, removing planning reduces success from 7/97/9 to 4/94/9, while removing retained experience reduces it to 3/93/9. Execution traces illustrate feedback-guided corrections and cross-episode reuse of an inspection strategy. Execution analyses illustrate how task structure and experience-derived feedback guide task progression, action correction, and experience reuse. These results support context-space adaptation as a practical mechanism for turning physical interaction into reusable guidance for robotic manipulation.

Figures & tables

Explore similar work

CardsList
  1. GPT-6-Astra Understands Touch and Outperform Vision--Tactile--Language--Action Models

    Oct 20, 2026Ziyong Wang, Yupei Lin, Runxuan Huang +3Tactile-Vision-Language-Action ModelsRobot Policy Learning

  2. Large Multimodal Models for Direct Robot Control: GPT-6 as a Beginning

    Oct 20, 2026Youwei Zhou, Bohan Hou, Yankai Fu +10Robotic ManipulationZero-Shot Robotic Manipulation

  3. IMBench: A Benchmark for Intuitive Robotic Manipulation

    Jul 17, 2026Anurag Maurya, Sukhvansh Jain, Prajwal Avhad +9Contact-Rich Robotic ManipulationRobotic Manipulation