Research map / Reinforcement learning / Visual and representation RL
Task Planning for Visual Room Rearrangement under Partial Observability
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes a hierarchical task planner for embodied visual room rearrangement under partial observability that combines a commonsense-LLM-based Search Network for finding unseen objects with a planner for simultaneous search and rearrangement.
Problem
Embodied agents must search for unseen objects and rearrange them to reach a tidy room state using only visual input under partial observability.
Approach
A hierarchical task planner in which a Search Network uses commonsense knowledge from large language models to guide search for unseen objects, interleaved with rearrangement planning.
Closest papers by meaning
- Task Planning for Object Rearrangement in Multi-Room Environments (AAAI 2024 · similarity 0.93)
- A Simple Approach for Visual Room Rearrangement: 3D Mapping and Semantic Search (ICLR 2023 · similarity 0.90)
- A Category Agnostic Model for Visual Rearrangment (CVPR 2024 · similarity 0.87)
- REGNav: Room Expert Guided Image-Goal Navigation (AAAI 2025 · similarity 0.85)
- RDD: Retrieval-Based Demonstration Decomposer for Planner Alignment in Long-Horizon Tasks (NeurIPS 2025 · similarity 0.84)
- Entity-Centric Reinforcement Learning for Object Manipulation from Pixels (ICLR 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.