X-Planner: Event-Structured Task Planning for Embodied Intelligence
arXiv cs.AIen
arXiv:2609.25187v1 Announce Type: new Abstract: Task planning bridges high-level instructions and executable behavior in long-horizon manipulation, yet modern Vision-Language-Action (VLA) systems often leave this intermediate structure implicit. Existing chain-of-thought (CoT) planners also tend to rely on coarse task-level annotations or serialize long reasoning traces token by token. We present X-Planner, a planning front-end that addresses both the supervision and representation of embodied reasoning. Our planning data combine Ego, UMI, and teleoperation under a hierarchy granularity with source-dependent annotation depth. Takeover-time annotations and human-designed failures supervise on
This is a short summary published by AI Global Wire. The full article is owned and hosted by arXiv cs.AI — open it there to read it in full.
Read the full story at arXiv cs.AI- Forskning
Related AI news
- 4DGS-JEPA: Temporally Compositional Joint-Embedding Prediction for Dynamic Gaussian SplattingarXiv cs.AI · September 23, 2026
- Lean Pool: An AI-Maintained Archive of Formalized MathematicsarXiv cs.AI · September 23, 2026
- Making Agents More Consistent: Skills Should Form Habits for Repeat TasksarXiv cs.AI · September 23, 2026
- Real-Time Hand Gesture Recognition for OpenXR Using Transformer-Based Machine LearningarXiv cs.AI · September 23, 2026
- Recovering Agentic Sovereignty: Mitigating the Consensus Paradox via Contrastive Epistemic DecodingarXiv cs.AI · September 23, 2026
- Efficient Iterative Retrieval with Heterogeneous BatchingarXiv cs.AI · September 23, 2026