World2Mind: Cognition Toolkit for Allocentric Spatial Reasoning in Foundation Models

2026-03-10

AI Summary

Existing multimodal foundation models often rely on statistical shortcuts or are constrained by 2D perception, limiting their generalization in spatial reasoning. Inspired by biological mechanisms of spatial cognition, this work proposes a training-free cognitive toolkit that constructs a structured spatial cognitive map through 3D reconstruction and instance segmentation. It introduces an Allocentric Spatial Tree (AST), parameterized by ellipses, to encode geometric-topological priors and enable models to actively acquire target spatial knowledge. This approach achieves, for the first time, zero-shot 3D spatial reasoning capability in purely text-based models: by leveraging only the structured textual descriptions generated by the AST, these models approach the performance of state-of-the-art multimodal systems, improving reasoning accuracy by 5%–18% on models such as GPT-5.2.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

Abstract

Achieving robust spatial reasoning remains a fundamental challenge for current Multimodal Foundation Models (MFMs). Existing methods either overfit statistical shortcuts via 3D grounding data or remain confined to 2D visual perception, limiting both spatial reasoning accuracy and generalization in unseen scenarios. Inspired by the spatial cognitive mapping mechanisms of biological intelligence, we propose World2Mind, a training-free spatial intelligence toolkit. At its core, World2Mind leverages 3D reconstruction and instance segmentation models to construct structured spatial cognitive maps, empowering MFMs to proactively acquire targeted spatial knowledge regarding interested landmarks and routes of interest. To provide robust geometric-topological priors, World2Mind synthesizes an Allocentric-Spatial Tree (AST) that uses elliptical parameters to model the top-down layout of landmarks accurately. To mitigate the inherent inaccuracies of 3D reconstruction, we introduce a three-stage reasoning chain comprising tool invocation assessment, modality-decoupled cue collection, and geometry-semantics interwoven reasoning. Extensive experiments demonstrate that World2Mind boosts the performance of frontier models, such as GPT-5.2, by 5%~18%. Astonishingly, relying solely on the AST-structured text, purely text-only foundation models can perform complex 3D spatial reasoning, achieving performance approaching that of advanced multimodal models.

Problem

• spatial reasoning
• multimodal foundation models
• 3D grounding
• generalization
• allocentric representation

Innovation

• allocentric spatial reasoning
• spatial cognitive mapping
• training-free toolkit
• geometry-semantics reasoning
• multimodal foundation models