PhySGA: A Physics-Space Grounded Agentic Framework for Generalizable and Interactive 3D Composition

arXiv:2610.20560 · cs.AI, cs.CV · Submitted 2026-10-14 · Read on arXiv

cs.AI, cs.CV

Submitted: 2026-10-14

Updated: 2026-10-14

Comments: Project Page: https://agenticei.github.io/PhySGA-page/

Project page: https://agenticei.github.io/PhySGA-page

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

Terminology

• Agentic Framework — 1× in this paper · explained in 6 episodes

Sources

• ReplicateAnyScene: Zero-Shot Video-to-3D Composition via Textual-Visual-Spatial Alignment
• SimuScene: Simulation-Ready Compositional 3D Scene Reconstruction from a Single Image
• SIRI-Bench: Challenging VLMs' Spatial Intelligence through Complex Reasoning Tasks
• OnlineX: Unified Online 3D Reconstruction and Understanding with Active-to-Stable State Evolution
• SkillRL: Evolving Agents via Recursive Skill-Augmented Reinforcement Learning

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