Hylos: Operability Contracts for Model-Native Spatial Intelligence
The paper titled 'Hylos: Operability Contracts for Model-Native Spatial Intelligence' introduces a new systems architecture aimed at enhancing the operability of generated 3D objects and environments. It emphasizes the importance of not only visual quality but also the reliability of 3D outputs for various applications such as CAD and robotics. The architecture includes mechanisms for managing spatial changes and ensuring that generated content can be effectively utilized in practical scenarios.
- ▪Hylos is designed to maintain operability state over various spatial elements.
- ▪The architecture routes durable spatial changes through a mechanism called SpatialTransaction.
- ▪The study highlights the need for spatial AI to be evaluated on its reliability for practical applications.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Artificial Intelligence arXiv:2605.24728 (cs) [Submitted on 23 May 2026] Title:Hylos: Operability Contracts for Model-Native Spatial Intelligence Authors:Christopher Da Silva View a PDF of the paper titled Hylos: Operability Contracts for Model-Native Spatial Intelligence, by Christopher Da Silva View PDF HTML (experimental) Abstract:Foundation models can increasingly describe, reconstruct, and generate 3D objects, assemblies, scenes, and environments, but visually plausible spatial output is not yet operable 3D. A generated object or environment becomes useful to an agent only when the system can identify its entities, frames, surfaces, constraints, provenance, admissible actions, expected effects, and validation failures.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
