Visual Prompt Guided Unified Pushing Policy
Not on the map. Why?
Abstract (arXiv)
As one of the simplest non-prehensile manipulation skills, pushing has been widely studied as an effective means to rearrange objects. Existing approaches, however, typically rely on multi-step push plans composed of pre-defined pushing primitives with limited application scopes, which restrict their efficiency and versatility across different scenarios. In this work, we propose a unified pushing policy that incorporates a lightweight prompting mechanism into a flow matching policy to guide the generation of reactive, multimodal pushing actions. The visual prompt can be specified by a high-level planner, enabling the reuse of the pushing policy across a wide range of planning problems. Experimental results demonstrate that the proposed unified pushing policy not only outperforms existing baselines but also effectively serves as a low-level primitive within a VLM-guided planning framework to solve table-cleaning tasks efficiently.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 24 of its 35 references that are in the map. First posted on arXiv 22 Feb 2026.
Builds on
24 of 35 references.
References · 24
- 2023Segment Anythingcited by 8,158
- 2022Scalable Diffusion Models with Transformerscited by 5,931
- 2022Classifier-Free Diffusion Guidancecited by 5,717
- 2022Flow Matching for Generative Modelingcited by 5,426
- 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusioncited by 2,730
- 2025Qwen3-VL Technical Reportcited by 2,534
- 2023Goal-Conditioned Imitation Learning using Score-based Diffusion Policiescited by 229
- 2024PIVOT: Iterative Visual Prompting Elicits Actionable Knowledge for VLMscited by 187
- 2022What Matters in Language Conditioned Robotic Imitation Learning over Unstructured Datacited by 137
- 2024Multimodal Diffusion Transformer: Learning Versatile Behavior from Multimodal Goalscited by 131
- 2024MOKA: Open-World Robotic Manipulation through Mark-Based Visual Promptingcited by 126
- 2018Learning Synergies between Pushing and Grasping with Self-supervised Deep Reinforcement Learningcited by 121
12 more
- 2024RT-Sketch: Goal-Conditioned Imitation Learning from Hand-Drawn Sketchescited by 41
- 2024Flow Matching Imitation Learning for Multi-Support Manipulationcited by 29
- 2025Robotic Visual Instructioncited by 21
- 2019A Deep Learning Approach to Grasping the Invisiblecited by 18
- 2016Feedback Control of the Pusher-Slider System: A Story of Hybrid and Underactuated Contact Dynamicscited by 13
- 2019Let's Push Things Forward: A Survey on Robot Pushingcited by 13
- 2019Robot Learning of Shifting Objects for Grasping in Cluttered Environmentscited by 12
- 2020DIPN: Deep Interaction Prediction Network with Application to Clutter Removalcited by 7
- 2017Learning to Singulate Objects using a Push Proposal Networkcited by 5
- 2025Visual Prompting for Robotic Manipulation with Annotation-Guided Pick-and-Place Using ACTcited by 5
- 2024UNO Push: Unified Nonprehensile Object Pushing via Non-Parametric Estimation and Model Predictive Controlcited by 4
- 2024Object-Centric Kinodynamic Planning for Nonprehensile Robot Rearrangement Manipulationcited by 2
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2025Push Smarter, Not Harder: Hierarchical RL-Diffusion Policy for Efficient Nonprehensile Manipulation
- 2025AdaptPNP: Integrating Prehensile and Non-Prehensile Skills for Adaptive Robotic Manipulation
- 2025Collaborative Multi-Robot Non-Prehensile Manipulation via Flow-Matching Co-Generation
- 2025AntiGrounding: Executable Robot Trajectories as Visual Prompts for VLM-Guided Manipulation
- 2026HiVLA: A Visual-Grounded-Centric Hierarchical Embodied Manipulation System
- 2026Leveraging Inter-object Affordances for Efficient Planning in Contact-rich Tasks
- 2025Gentle Manipulation Policy Learning via Demonstrations from VLM Planned Atomic Skills ↗
- 2026Zero-shot Interactive Perception