World2Mind: Cognition Toolkit for Allocentric Spatial Reasoning in Foundation Models

TL;DR

World2Mind introduces a spatial reasoning toolkit that constructs structured 3D cognitive maps, enabling foundation models to perform robust allocentric spatial reasoning without additional training, significantly improving their accuracy and generalization.

Contribution

It presents a training-free spatial reasoning framework using 3D reconstruction and structured maps, enhancing foundation models' spatial understanding capabilities.

Findings

Boosts GPT-5.2 performance by 5-18%.

Purely text-based models approach multimodal model performance in spatial reasoning.

Enables models to perform complex 3D reasoning using structured spatial maps.

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.