S-Agent: Spatial Tool-Use Elicits Reasoning for Spatial Intelligence
Real-world spatial intelligence requires reasoning over a continuous and evolving 3D world, yet existing VLMs and too...
Source Evidence
What Changed
Real-world spatial intelligence requires reasoning over a continuous and evolving 3D world, yet existing VLMs and too...
Why It Matters
S‑Agent shows that spatial reasoning can be built on continuous evidence accumulation rather than static snapshots, enabling even 8‑B parameter models to match or surpass commercial giants on multi‑view and video tasks without costly training. This turns perception into dynamic scene‑centric planning, widening the bandwidth of autonomous agents for navigation, robotics, and immersive VR where real‑time 3‑D understanding is critical.
Confirmed Facts
Real-world spatial intelligence requires reasoning over a continuous and evolving 3D world, yet existing VLMs and tool-augmented agents largely remain tied to static, stateless inference from isolated visual observations. We introduce \textsc{S-Agent}, a spatial tool-use agentic paradigm for understanding and reasoning over continuous multi-view images and videos. By formulating spatial reasoning as spatio-temporal evidence accumulation rather than isolated frame-level prediction, S-Agent reshapes spatial perception into scene-centric understanding beyond frame-centric recognition. Specifically, S-Agent casts the VLM as a semantic planner that decides what evidence is needed, while a hierarchy of spatial tools and experts grounds objects in 2D, lifts them into 3D geometric evidence, and aggregates this evidence into high-level spatial knowledge (e.g., counting, measurement, orientation, and relative position). Additionally, a temporal memory mechanism, including Scene Memory for maintaining the evolving scene state and Agent Memory for accumulating reasoning context, enables evidence integration across frames and reasoning steps. Comprehensive experiments on multi-view and video spatial reasoning benchmarks show that S-Agent consistently improves both open-source and closed-source VLMs in a training-free manner. Beyond inference-time augmentation, supervised fine-tuning (SFT) on S-Agent-generated spatial trajectories S-300K yields S-Agent-8B, a compact spatial agent that significantly surpasses similar-scale baselines (e.g., Qwen3-VL-8B) and performs comparably to advanced closed-source models (e.g., GPT-5.4 and Gemini 3).
Who Is Affected
- Qwen
- Google DeepMind
- GPT
- AI product teams
What To Watch Next
- Watch for benchmark validation, API availability, pricing, limits, and early customer adoption.
- Watch whether additional sources confirm the same claim.
You will be redirected to huggingface.co.