Spatial-Interactor: Learning Spatial Reasoning through Interaction with the Observable Physical World

arXiv:2609.23038 · cs.AI, cs.LG · Submitted 2026-09-19 · Read on arXiv

cs.AI, cs.LG

Submitted: 2026-09-19

Updated: 2026-10-12

Project page: https://zju-omniai.github.io/Spatial-Interactor

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

Terminology

• Distillation — 2× in this paper · explained in 2 episodes
• Benchmark — 1× in this paper · explained in 2 episodes
• Fine-Tuning — 1× in this paper · explained in 5 episodes
• On-Policy Distillation — 1× in this paper · explained in 3 episodes
• Self-distillation — 1× in this paper · explained in 6 episodes
• Supervised Fine-Tuning (SFT) — 1× in this paper · explained in 8 episodes
• Two-Stage Training Strategy — 1× in this paper · explained in 3 episodes
• Two-stage training — 1× in this paper · explained in 2 episodes
• Vision-Language Models (VLMs) — 1× in this paper · explained in 6 episodes

Sources

• Qwen3-VL Technical Report
• Qwen2.5-VL Technical Report
• SIMS-V: Simulated Instruction-Tuning for Spatial Video Understanding
• SpatialDreamer: Incentivizing Spatial Reasoning via Active Mental Imagery
• Holi-Spatial: Evolving Video Streams into Holistic 3D Spatial Intelligence
• Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
• ESI-Bench: Towards Embodied Spatial Intelligence that Closes the Perception-Action Loop
• Uni-OPD: Unifying On-Policy Distillation with a Dual-Perspective Recipe
• Reinforcement Learning via Self-Distillation
• AI2-THOR: An Interactive 3D Environment for Visual AI
• ViewSpatial-Bench: Evaluating Multi-perspective Spatial Localization in Vision-Language Models
• Visual-OPSD: Cross-Modal On-Policy Self-Distillation for Efficient Unified Multimodal Reasoning
• SpaMEM: Benchmarking Dynamic Spatial Reasoning via Perception-Memory Integration in Embodied Environments
• GPT-4o System Card
• SpaceR: Reinforcing MLLMs in Video Spatial Reasoning
• Qwen3.5-Omni Technical Report
• The Replica Dataset: A Digital Replica of Indoor Spaces
• Planning with the Views
• Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation
• World2VLM: Distilling World Model Imagination into VLMs for Dynamic Spatial Reasoning

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