3 papers
cs.AI2026
ASH: Agents that Self-Hone via Embodied Learning
Benjamin Schneider, Xavier Schneider, Victor Zhong +1
Long-horizon embodied tasks remain a fundamental challenge in AI, as current methods rely on hand-engineered rewards or action-labeled demonstrations, neither of which scales. We i…
cs.RO2025
Symskill: Symbol and Skill Co-Invention for Data-Efficient and Reactive Long-Horizon Manipulation
Yifei Simon Shao, Yuchen Zheng, Sunan Sun +3
Multi-step manipulation in dynamic environments remains challenging. Imitation learning (IL) is reactive but lacks compositional generalization, since monolithic policies do not de…
cs.RO2025
Elastic Motion Policy: An Adaptive Dynamical System for Robust and Efficient One-Shot Imitation Learning
Tianyu Li, Sunan Sun, Shubhodeep Shiv Aditya +1
Behavior cloning (BC) has become a staple imitation learning paradigm in robotics due to its ease of teaching robots complex skills directly from expert demonstrations. However, BC…