SPHERE: Adaptive VR Indoor Scene Generation via LLM-Enhanced Spatial Preference Learning and Human-in-the-Loop RL

arXiv:2610.02023 · cs.AI, cs.HC · Submitted 2026-10-01 · Read on arXiv

cs.AI, cs.HC

Submitted: 2026-10-01

Updated: 2026-10-01

Code: https://github.com/hyeonmin11/SPHERE

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Human-in-the-Loop — 1× in this paper · explained in 3 episodes

Sources

• GPT-4 Technical Report
• Open-Universe Indoor Scene Generation using LLM Program Synthesis and Uncurated Object Databases
• PREDICT: Preference Reasoning by Evaluating Decomposed preferences Inferred from Candidate Trajectories
• AutoGPT+P: Affordance-based Task Planning with Large Language Models
• SceneSeer: 3D Scene Design with Natural Language
• LLF-Bench: Benchmark for Interactive Learning from Language Feedback
• SceneNAT: Masked Generative Modeling for Language-Guided Indoor Scene Synthesis
• Direct Language Model Alignment from Online AI Feedback
• SceneLinker: Compositional 3D Scene Generation via Semantic Scene Graph from RGB Sequences
• Towards General Text Embeddings with Multi-stage Contrastive Learning
• CLIP-Layout: Style-Consistent Indoor Scene Synthesis with Semantic Furniture Embedding
• LLM-Powered Hierarchical Language Agent for Real-time Human-AI Coordination
• DiffusER: Discrete Diffusion via Edit-based Reconstruction
• Conversational User-AI Intervention: A Study on Prompt Rewriting for Improved LLM Response Generation
• RAGEN: Understanding Self-Evolution in LLM Agents via Multi-Turn Reinforcement Learning
• Monte Carlo Tree Search Boosts Reasoning via Iterative Preference Learning
• LLplace: The 3D Indoor Scene Layout Generation and Editing via Large Language Model
• Learning Structural Edits via Incremental Tree Transformations
• Learning to Represent Edits
• Fine-Tuning Language Models from Human Preferences

Related papers

• MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
• Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
• Agents' Last Exam
• The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
• MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
• Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems