[2603.01465] Non-Markovian Long-Horizon Robot Manipulation via Keyframe Chaining

Computer Science > Robotics arXiv:2603.01465 (cs) [Submitted on 2 Mar 2026] Title:Non-Markovian Long-Horizon Robot Manipulation via Keyframe Chaining Authors:Yipeng Chen, Wentao Tan, Lei Zhu, Fengling Li, Jingjing Li, Guoli Yang, Heng Tao Shen View a PDF of the paper titled Non-Markovian Long-Horizon Robot Manipulation via Keyframe Chaining, by Yipeng Chen and 6 other authors View PDF HTML (experimental) Abstract:Existing Vision-Language-Action (VLA) models often struggle to generalize to long-horizon tasks due to their heavy reliance on immediate observations. While recent studies incorporate retrieval mechanisms or extend context windows to handle procedural tasks, they often struggle to capture Non-Markovian dependencies, where optimal actions rely solely on specific past states rather than the current observation. To address this, we introduce Keyframe-Chaining VLA, a framework that extracts and links key historical frames to model long-horizon dependencies. Specifically, we propose an automatic keyframe selector that learns a discriminative embedding space, effectively identifying distinct state transitions. To capture task-critical information, we design a progress-aware query mechanism that dynamically retrieves historical frames based on their temporal relevance to the current execution phase. These selected keyframes are integrated into the VLA as interleaved visual tokens, explicitly grounding the policy in the long-horizon temporal context. Finally, we introduce...