CoRe3D: Collaborative Reasoning as a Foundation for 3D Intelligence

TL;DR

CoRe3D presents a unified reasoning framework for 3D understanding and generation, integrating semantic and spatial reasoning to improve alignment, interpretability, and content fidelity in 3D models guided by language.

Contribution

It introduces a novel spatially grounded reasoning approach that decomposes 3D space, enabling high-level language-driven content formation with enhanced local consistency.

Findings

Produces 3D outputs with strong local consistency.

Achieves faithful alignment with linguistic descriptions.

Demonstrates effective reasoning over geometry and semantics.

Abstract

Recent advances in large multimodal models suggest that explicit reasoning mechanisms play a critical role in improving model reliability, interpretability, and cross-modal alignment. While such reasoning-centric approaches have been proven effective in language and vision tasks, their extension to 3D remains underdeveloped. CoRe3D introduces a unified 3D understanding and generation reasoning framework that jointly operates over semantic and spatial abstractions, enabling high-level intent inferred from language to directly guide low-level 3D content formation. Central to this design is a spatially grounded reasoning representation that decomposes 3D latent space into localized regions, allowing the model to reason over geometry in a compositional and procedural manner. By tightly coupling semantic chain-of-thought inference with structured spatial reasoning, CoRe3D produces 3D outputs that exhibit strong local consistency and faithful alignment with linguistic descriptions.