paper-with-me

홈 › Papers

CodeGraphVLP: Code-as-Planner Meets Semantic-Graph State for Non-Markovian Vision-Language-Action Models

2026-04-24 · Khoa Vo, Sieu Tran, Taisei Hanyu, Yuki Ikebe, Duy Nguyen, Nghi D. Q. Bui, Minh Vu, Anthony Gunderman, Chase Rainwater, Anh Nguyen, Ngan Le arxiv

Vision-Language-Action (VLA) models promise generalist robot manipulation, but are typically trained and deployed as short-horizon policies that assume the latest observation is sufficient for action reasoning. This assumption breaks in non-Markovian long-horizon tasks, where task-relevant evidence can be occluded or appear only earlier in the trajectory, and where clutter and distractors make fine-grained visual grounding brittle. We present CodeGraphVLP, a hierarchical framework that enables reliable long-horizon manipulation by combining a persistent semantic-graph state with an executable code-based planner and progress-guided visual-language prompting. The semantic-graph maintains task-relevant entities and relations under partial observability. The synthesized planner executes over this semantic-graph to perform efficient progress checks and outputs a subtask instruction together with subtask-relevant objects. We use these outputs to construct clutter-suppressed observations that focus the VLA executor on critical evidence. On real-world non-Markovian tasks, CodeGraphVLP improves task completion over strong VLA baselines and history-enabled variants while substantially lowering planning latency compared to VLM-in-the-loop planning. We also conduct extensive ablation studies to confirm the contributions of each component.

📄 PDF Abstract BibTeX arXiv:2604.22238

Code (0)

등록된 구현이 없습니다.

Tasks

Robot ManipulationVisual Grounding

Similar Papers 제목 키워드 기반

Situationally-aware Path Planning Exploiting 3D Scene Graphs

2025-08-08 · Saad Ejaz, Marco Giberna, Muhammad Shaheer, Jose Andres Millan-Romera 외 arxiv

3D Scene Graphs integrate both metric and semantic information, yet their structure remains underutilized for improving path planning efficiency and interpretability. In this work, we present S-Path, a situationally-awar…

Planner3D: LLM-enhanced graph prior meets 3D indoor scene explicit regularization

2024-03-19 · Yao Wei, Martin Renqiang Min, George Vosselman, Li Erran Li 외

Compositional 3D scene synthesis has diverse applications across a spectrum of industries such as robotics, films, and video games, as it closely mirrors the complexity of real-world multi-object environments. Convention…

3D Shape GenerationDiversityLanguage ModelingLanguage Modelling+3

GraphPlanner: Graph Memory-Augmented Agentic Routing for Multi-Agent LLMs

2026-04-26 · Tao Feng, Haozhen Zhang, Zijie Lei, Peixuan Han 외 arxiv

LLM routing has achieved promising results in integrating the strengths of diverse models while balancing efficiency and performance. However, to support more realistic and challenging applications, routing must extend i…

Computational EfficiencyReinforcement Learning

A Real-time Spatio-Temporal Trajectory Planner for Autonomous Vehicles with Semantic Graph Optimization

2025-02-25 · Shan He, Yalong Ma, Tao Song, Yongzhi Jiang 외

Planning a safe and feasible trajectory for autonomous vehicles in real-time by fully utilizing perceptual information in complex urban environments is challenging. In this paper, we propose a spatio-temporal trajectory …

Autonomous VehiclesBenchmarkingTrajectory Planning

When Heterophily Meets Heterogeneous Graphs: Latent Graphs Guided Unsupervised Representation Learning

2024-09-01 · Zhixiang Shen, Zhao Kang

Unsupervised heterogeneous graph representation learning (UHGRL) has gained increasing attention due to its significance in handling practical graphs without labels. However, heterophily has been largely ignored, despite…

Graph Representation LearningRepresentation Learning