{"ID":2870072,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.14460","arxiv_id":"2509.14460","title":"Learning Discrete Abstractions for Visual Rearrangement Tasks Using Vision-Guided Graph Coloring","abstract":"Learning abstractions directly from data is a core challenge in robotics. Humans naturally operate at an abstract level, reasoning over high-level subgoals while delegating execution to low-level motor skills -- an ability that enables efficient problem solving in complex environments. In robotics, abstractions and hierarchical reasoning have long been central to planning, yet they are typically hand-engineered, demanding significant human effort and limiting scalability. Automating the discovery of useful abstractions directly from visual data would make planning frameworks more scalable and more applicable to real-world robotic domains. In this work, we focus on rearrangement tasks where the state is represented with raw images, and propose a method to induce discrete, graph-structured abstractions by combining structural constraints with an attention-guided visual distance. Our approach leverages the inherent bipartite structure of rearrangement problems, integrating structural constraints and visual embeddings into a unified framework. This enables the autonomous discovery of abstractions from vision alone, which can subsequently support high-level planning. We evaluate our method on two rearrangement tasks in simulation and show that it consistently identifies meaningful abstractions that facilitate effective planning and outperform existing approaches.","short_abstract":"Learning abstractions directly from data is a core challenge in robotics. Humans naturally operate at an abstract level, reasoning over high-level subgoals while delegating execution to low-level motor skills -- an ability that enables efficient problem solving in complex environments. In robotics, abstractions and hie...","url_abs":"https://arxiv.org/abs/2509.14460","url_pdf":"https://arxiv.org/pdf/2509.14460v2","authors":"[\"Abhiroop Ajith\",\"Constantinos Chamzas\"]","published":"2025-09-17T22:25:06Z","proceeding":"cs.RO","tasks":"[\"cs.RO\"]","methods":"[]","has_code":false}