Generalization in Robotic Manipulation

Overview

The field studies whether robots can manipulate objects reliably when their shape, position, appearance, surroundings, or physical properties differ from training, and whether skills transfer across tasks and robot bodies. Common approaches use object-centric geometry and affordances, reusable skills and action primitives, symmetry-aware policies, and vision-language-action or world models that connect instructions and predicted outcomes to control. Researchers also expand training through simulation, human videos, data augmentation, and diverse demonstrations, then test policies in settings such as LIBERO, RLBench, and real-world object manipulation. A central challenge is that high benchmark success does not always mean a robot has learned physical or compositional generalization; evaluations must distinguish genuine transfer from memorized patterns and expose failures under combined shifts.

Latest papers

CardsList

381 papers31 papers in the last four weeks, down 47% on the four weeks before. 0.2% of all new papers.

  1. Affordance Transfer Across Object Instances via Semantically Anchored Functional Map

    Feb 16, 2026Xiaoxiang Dong, Weiming ZhiFunctional Maps3D Shape Matching

  2. RoboAug: One Annotation to Hundreds of Scenes via Region-Contrastive Data Augmentation for Robotic Manipulation

    Feb 15, 2026Xinhua Wang, Kun Wu, Zhen Zhao +10Synthetic Data AugmentationGeneralization in Robotic Manipulation

  3. HuMI: Humanoid Whole-Body Manipulation from Humanoid-Free Demonstrations

    Feb 6, 2026Ruiqian Nai, Boyuan Zheng, Junming Zhao +8Whole-Body Robotic ManipulationLearning from Demonstration

  4. STORM: Slot-based Task-aware Object-centric Representation for robotic Manipulation

    Jan 28, 2026Alexandre Chapin, Emmanuel Dellandréa, Liming ChenVision Foundation Model AdaptationGeneralization in Robotic Manipulation

  5. Sim-and-Human Co-training for Data-Efficient and Scene-Generalizable Bimanual Manipulation

    Jan 27, 2026Kaipeng Fang, Weiqing Liang, Yuyang Li +5Robot Skill LearningBimanual Robotic Manipulation

  6. Skill-Aware Diffusion for Generalizable Robotic Manipulation

    Jan 16, 2026Aoshen Huang, Jiaming Chen, Jiyu Cheng +3Robotic ManipulationGeneralization in Robotic Manipulation

  7. Where to Touch, How to Contact: A Hierarchical RL-MPC Framework for Geometry-Aware Sim-to-Real Manipulation

    Jan 16, 2026Zhixian Xie, Yu Xiang, Michael Posa +1Contact-Rich ManipulationNon-Prehensile Manipulation

  8. Learning Generalizable Hand-Object Tracking from Synthetic Demonstrations

    Dec 22, 2025Runyi Yu, Xiaoyi Lin, Hok Wai Tsui +10Robotic ManipulationGeneralization in Robotic Manipulation

  9. Learning Semantic Atomic Skills for Multi-Task Robotic Manipulation

    Dec 20, 2025Yihang Zhu, Weiqing Wang, Shijie Wu +2Long-Horizon Robotic ManipulationRobot Skill Learning

  10. Guardian: Detecting Robotic Manipulation Failures with Vision-Language Models

    Dec 1, 2025Paul Pacaud, Ricardo Garcia, Shizhe Chen +1Failure Detection for VLAsRobot Manipulation Datasets

  11. METIS: Multi-Source Egocentric Training for Integrated Dexterous Vision-Language-Action Model

    Nov 21, 2025Yankai Fu, Ning Chen, Junkai Zhao +5Egocentric Video Learning for RoboticsRobotic Manipulation

  12. AFFORD2ACT: How Far Can Sparse Keypoints Go for Robot Manipulation?

    Oct 1, 2025Anukriti Singh, Kasra Torshizi, Khuzema Habib +3Robotic ManipulationData-Efficient Robot Learning

  13. RoboPilot: Generalizable Dynamic Robotic Manipulation with Dual-thinking Modes

    Sep 30, 2025Xinyi Liu, Mohammadreza Fani Sani, Zewei Zhou +3Robotic ManipulationDynamic Robotic Manipulation

  14. BiNoMaP: Learning Category-Level Bimanual Non-Prehensile Manipulation Primitives

    Sep 25, 2025Huayi Zhou, Kui JiaNon-Prehensile Robotic ManipulationBimanual Robotic Manipulation

  15. EquiContact: A Hierarchical SE(3) Vision-to-Force Equivariant Framework for Spatially Generalizable Contact-rich Tasks

    Jul 15, 2025Joohwan Seo, Arvind Kruthiventy, Soomi Lee +5Contact-Rich Robotic ManipulationRobot Learning

  16. Data Scaling Laws in Imitation Learning for Robotic Manipulation

    Oct 24, 2024Fanqi Lin, Yingdong Hu, Pingyue Sheng +3Generalization in Robotic ManipulationRobotic Manipulation

  17. Learning Beyond Rigid Manipulation with Hybrid Soft-Rigid Robots

    Oct 10, 2024Mariano Ramírez Montero, Ebrahim Shahabi, Giovanni Franzese +3Robotic ManipulationSoft Robotics